CN115952887A - Event prediction method and information processing apparatus applied to event prediction - Google Patents
Event prediction method and information processing apparatus applied to event prediction Download PDFInfo
- Publication number
- CN115952887A CN115952887A CN202211548149.4A CN202211548149A CN115952887A CN 115952887 A CN115952887 A CN 115952887A CN 202211548149 A CN202211548149 A CN 202211548149A CN 115952887 A CN115952887 A CN 115952887A
- Authority
- CN
- China
- Prior art keywords
- event
- prediction
- events
- probability
- predicted
- 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
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
There is provided, in accordance with an embodiment of the present disclosure, a method of event prediction, including: the method comprises the steps of receiving one or more observation data related to a predicted event, analyzing the one or more observation data based on an event prediction model and outputting an event grade of the occurrence of the predicted event, wherein the event prediction model comprises a hierarchical analysis network, the hierarchical analysis network comprises a plurality of types of prior events, each type of prior event comprises one or more characterization events, and each type of prior event corresponds to the event grade of the occurrence of the predicted event. The embodiment of the disclosure further provides an information processing device for event prediction, which includes a processor configured to execute the event prediction method.
Description
Technical Field
Example embodiments of the present disclosure generally relate to the field of computer information, and in particular, to an event prediction method and an information processing apparatus for event prediction.
Background
Event prediction is applied to various fields of industry and industry, such as power, network fault prediction, energy distribution prediction, production decision, transportation, environmental monitoring, military and the like.
However, the existing event prediction methods only simply rely on the collected monitoring data and the historical information, and the relationships between the data and between the historical information are not well comprehensively considered, so that the current event prediction capability is insufficient.
Disclosure of Invention
In a first aspect of the present disclosure, there is provided a method of event prediction, comprising: the method comprises the steps of receiving one or more observation data related to a predicted event, analyzing the one or more observation data based on an event prediction model and outputting an event grade of the occurrence of the predicted event, wherein the event prediction model comprises a hierarchical analysis network, the hierarchical analysis network comprises a plurality of types of prior events, each type of prior event comprises one or more characterization events, and each type of prior event corresponds to the event grade of the occurrence of the predicted event.
In a second aspect of the present disclosure, there is provided an information processing apparatus for event prediction, comprising a processor configured to execute the above event prediction method.
According to the event prediction method and the information processing device provided by the embodiment of the invention, the event prediction model constructed by the hierarchical analysis network is utilized, and the degree of the event to which the event belongs is predicted by analyzing observation data systematically layer by layer and step by step of a plurality of classes of prior events, characterization events corresponding to each class of prior events and a plurality of event grades corresponding to each class of prior events, so that the prediction precision and the breadth of the event are improved.
It should be understood that the statements herein set forth in this summary are not intended to limit the essential or critical features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description.
Drawings
The above and other features, advantages and aspects of various embodiments of the present disclosure will become more apparent by referring to the following detailed description when taken in conjunction with the accompanying drawings. Wherein:
fig. 1 is a flowchart illustrating an event prediction method according to an embodiment of the disclosure.
Fig. 2 is a schematic diagram illustrating a model architecture of an event prediction model used in an event prediction method according to an embodiment of the present disclosure.
Detailed Description
Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the present disclosure are illustrated in the accompanying drawings, it is to be understood that the present disclosure may be embodied in various forms and should not be construed as limited to the embodiments set forth herein, but rather are provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and the embodiments of the disclosure are for illustration purposes only and are not intended to limit the scope of the disclosure.
Referring to fig. 1, an embodiment of the invention provides a method 10 for event prediction, where the method 10 includes:
s12, receiving one or more observations relating to said predicted event, an
And S14, analyzing the one or more observation data based on an event prediction model and outputting an event grade of the occurrence of the predicted event, wherein the event prediction model comprises a hierarchical analysis network, the hierarchical analysis network comprises a plurality of types of prior events, each type of prior event comprises one or more characterization events, and each type of prior event corresponds to the event grade of the occurrence of the predicted event.
The event prediction method 10 can be applied to various technical fields, such as power, network failure prediction, energy distribution prediction, production decision, transportation, environmental monitoring, military and the like. The event prediction can be fault prediction, natural disaster risk assessment, energy distribution prediction and the like.
In step S12, the one or more observation data may be obtained according to an application scenario of the predicted event, and the one or more observation data may include one or more actually occurring observation events, specific information or data of the observation events, and time information of the occurrence of the observation events. The one or more observation data may be obtained by an information acquisition device according to an application scenario. The information collecting device may be, but is not limited to, one or more of a sensor, a detector, and a network communication device. The one or more observed data may reflect a particular condition of the observed event to predict a likelihood of a future event occurring based on the particular condition.
In step S14, the event prediction model is used to determine an event level of the predicted event occurrence by using the input one or more of the observation data.
Referring to fig. 1 and 2 together, the event prediction method 10 further includes: and S11, constructing the event prediction model.
The event prediction model comprises a hierarchical analysis network constructed by a hierarchical analysis method. The step S11 of constructing the event prediction model further includes:
s112, constructing a plurality of event levels for each predicted event;
s114, aiming at each event grade, constructing multiple types of prior events;
s116, and constructing a plurality of characterization events for each type of prior event, an
And S118, determining the probability of the occurrence of the predicted event based on the characterization event, the prior event and the predicted probability of the event grade.
In an embodiment of the present disclosure, a plurality of event levels X may be constructed for each predicted event in the event prediction model i For example, the first three event levels, the first three, the obvious, and the predicted probability of each event level can be represented as X 1 ，X 2 ，X 3 Range of values thereofAre respectively set as [0,s i ]I =1,2,3. Wherein s is i Different values may be set depending on the event class. The first three distinct event levels represent different degrees and likelihoods of possible occurrence of the predicted event. When the primary representation corresponds to the characterization event, a predicted event may occur, when the primary representation clearly indicates that the corresponding characterization event occurs, a predicted event may occur with a high probability, and when the primary representation directly indicates that the corresponding characterization event occurs, the predicted event may be basically determined to occur. It is to be understood that the event levels described in the embodiments of the present disclosure are not limited to three.
In one embodiment of the present disclosure, the event level X i Including m-class prior events Y associated with the predicted event ij J is more than or equal to 1 and less than or equal to m. The m-class prior event Y ij Can be obtained by comprehensive analysis according to expert knowledge or historical data. The prior event Y ij The prediction probability value range of (2) is [0,q ] ij ]。
In one embodiment of the present disclosure, the prior event Y ij May include n specific characterization events Z ijk And k is greater than or equal to 1 and less than or equal to n, the prediction method 100 can further set the characterization event Z according to historical data ijk Based on the prediction probability
Time effect parameter->
And time effect decay functions and their parameters.
In one embodiment of the present disclosure, building the event prediction model may further configure a parameter that reflects whether the predicted event has an impact at a certain time. In particular, the parameter may be a constant ε when calculating T ijk Characterising events Z occurring at a time ijk Prediction probability P at time T ijk When (T) is equal to or greater than P ijk (T)<ε then considers the characterization event Z ijk Without influence at time T, in other words, the characterizing event Z ijk For time TThe prediction of the predicted event does not work.
In an embodiment of the present disclosure, the step S14 further includes:
s142, determining one or more first characterization events associated with the one or more observed data based on the one or more observed data;
s144, determining one or more first prior events corresponding to the first characterization event, an
S146, determining an event grade of the occurrence of the predicted event based on the one or more first prior events.
In one embodiment of the present disclosure, the step S142 further includes:
s1422, determining a prediction probability of said one or more characterizing events at a prediction time based on said one or more observations, an
S1424, determining one or more of the first characterization events based on the prediction probability.
In one embodiment of the present disclosure, the event Z may be characterized according to the observation data ijk Based on the prediction probability
Current time T, event occurrence time T ijk Time effect decay function and parameters thereof to calculate a characterizing event Z ijk Predicted probability P at time T ijk (T). The time-effect decay function may be, but is not limited to, a gaussian function, and the time-effect-based decay function determination of the predicted probability of the one or more characterizing events at the predicted time instant:
said parameter
For the characterization event Z ijk Basic prediction probability ofNumber T ijk For the characterization event Z ijk The time of occurrence, the parameter->
For time-effect parameters, parameters A and σ are parameters of the time decay function, if P ijk (T)<Epsilon, then represents the characterization event Z ijk There is no effect at time T and prediction of the subsequent event level may not be used. The prediction probabilities of one or more of the characterizing events at the predicted time instant may be ranked from large to small for analysis of subsequent prior events.
In one embodiment of the present disclosure, the step S144 further includes:
s1442, determining a prediction probability at a prediction time for each type of the prior event based on the prediction probabilities at the prediction time for the one or more first characterized events, an
S1444, determining one or more first prior events based on the prediction probability of each type of the prior events at the prediction time.
In one embodiment of the present disclosure, the prediction probability of the prior event at the prediction time is obtained by the following calculation method:
p' ijk (T) is a prior event Y ij Prediction probability at a prediction time T, said prior event Y ij The prediction probability of (2) is in the range of [0,q ] ij ]If the event is a priori Y ij All of the characterization events in (1) have an effect at time T, then the prediction probability of YIj takes its upper bound q ij Otherwise, T moment prior event Y ij The predicted probability of (2) is given by the above P' ijk (T) formula calculation, the formula P' ijk (T) over n iterations, P' ijn (T) is the prior event Y at the moment T ij Predicted probability P' ij The parameter n is the prior event Y ij Of the plurality of characterization events. What is needed isCalculating a plurality of predicted probabilities P 'of interest' ij Can be arranged from large to small according to the value of the probability and is used for subsequently calculating the event grade X i The prediction probability of (2).
In one embodiment of the present disclosure, the step S146 further includes:
s1462, determining a predicted probability of each event class of occurrence of one or more first prior events based on the predicted probability of the one or more first prior events at the predicted time, an
And S1464, determining the probability of the event grade of the occurrence of the predicted event according to the prediction probability of each event grade.
In one embodiment of the present disclosure, the prediction probability of the event level is obtained by the following calculation method:
wherein the event rank X i Has a prediction probability of P " i (T), the prediction probability P " i (T) is in the range of [0,s i ]After m iterations, P ″) im (T) is the event level X at the time T i Is denoted as P ″) i The parameter m represents the number of the a priori events.
The probability P (T) of the event level of the predicted event occurrence is:
P(T)＝max(P″ i1 ,P″ i2 ,P″ i3 …)，
the parameters i1, i2, i3 represent the number of the event class.
In the embodiment of the disclosure, the event grade of the predicted event can be determined by obtaining the probability of the event grade of the predicted event, and the probability of the event to be predicted can be determined according to the grade of the predicted event.
In an embodiment of the present disclosure, the event prediction method 10 further includes obtaining an event level of the predicted event occurring at a specific time.
In an embodiment of the present disclosure, the event prediction method 10 further comprises generating one or more policy information based on a level of the predicted event occurrence.
The event prediction method provided by the embodiment of the disclosure is an event prediction method based on dynamic logic, and by using an event prediction model constructed by a hierarchical analysis network, the possibility of event occurrence can be predicted through qualitative and quantitative combined systematic and hierarchical analysis, specifically, the event class to which the event occurrence belongs is predicted by systematically analyzing observation data layer by layer and step by step of a plurality of classes of prior events, characterization events corresponding to each class of prior events and a plurality of event classes corresponding to each class of prior events, so that the prediction capability and precision of the event are improved.
An embodiment of the present invention further provides an information processing apparatus, where the information processing apparatus is configured to perform the event prediction, and the information processing apparatus at least includes a processor, which can execute all the steps in the event prediction method 10 described above, and is not described again. The information processing device may be any computer or other electronic device having data analysis. The information processing apparatus may further include a computer storage medium for storing the observation data and the event prediction model.
The media may be any available media that is accessible by the information processing device and includes, but is not limited to, volatile and non-volatile media, removable and non-removable media. The memory may be volatile memory (e.g., registers, cache, random Access Memory (RAM)), non-volatile memory (e.g., read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory), or some combination thereof. The storage devices may be removable or non-removable media, and may include machine-readable media, such as flash drives, diskettes, or any other media.
Various aspects of the present disclosure are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus, devices and computer program products implemented in accordance with the disclosure. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer-readable program instructions.
The computer readable program instructions may be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer, other programmable apparatus or other devices implement the functions/acts specified in the flowchart and/or block diagram block or blocks.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various implementations of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The foregoing has described implementations of the present disclosure, and the above description is illustrative, not exhaustive, and is not limited to the implementations disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described implementations. The terminology used herein was chosen in order to best explain the principles of various implementations, the practical application, or improvements to the technology in the marketplace, or to enable others of ordinary skill in the art to understand various implementations disclosed herein.
Claims (13)
1. A method of event prediction, comprising:
receiving one or more observations relating to the predicted event, an
Analyzing the one or more observation data based on an event prediction model and outputting an event grade of the occurrence of the predicted event, wherein the event prediction model comprises a hierarchical analysis network, the hierarchical analysis network comprises a plurality of types of prior events, each type of prior event comprises one or more characterization events, and each type of prior event corresponds to the event grade of the occurrence of the predicted event.
2. The event prediction method of claim 1, wherein analyzing the one or more observations based on the event prediction model further comprises: determining one or more first characterization events based on the one or more observations;
determining one or more first prior events corresponding to the first characterizing events, an
Determining an event rank of the predicted event occurrence based on the one or more first prior events.
3. The event prediction method of claim 2, wherein determining one or more of the first characterized events further comprises:
determining a prediction probability of the one or more characterizing events at a prediction time based on the one or more observations, an
Determining one or more of the first characterization events based on the prediction probability.
4. The event prediction method of claim 3, wherein the time-based effect decay function of the prediction probability of the one or more characterizing events at the prediction time determines:
wherein the event Z is characterized ijk The prediction probability at said prediction instant T is denoted P ijk (T), the parameter i corresponds to the grade, j corresponds to the jth prior event, k corresponds to the kth characterization event, and the parameter
For the characterization event Z ijk Based on the prediction probability, parameter T ijk For the characterization event Z ijk The time of occurrence, the parameter->
For time-effect parameters, parameters A and σ are parameters of the time decay function, if P ijk (T)<Epsilon, then represents the characterization event Z ijk At time T, the parameter epsilon is a constant, with no effect.
5. The event prediction method of any of claims 2-4, wherein determining one or more of the first prior events further comprises:
determining a prediction probability of each type of the prior event at a prediction time based on the prediction probabilities of the one or more first characterizing events at the prediction time, and determining one or more of the first prior events based on the prediction probability of each type of the prior event at the prediction time.
6. The event prediction method according to claim 5, wherein the prediction probability of the prior event at the prediction time is obtained by the following calculation method:
p' ijk (T) is an a priori event Y ij Prediction probability at a prediction instant T, said prior event Y ij The prediction probability of (2) is in the range of [0,q ] ij ]If the event is a priori Y ij All of the characterization events in (1) have an effect at time T, then Y ij Is taken to its upper limit q ij Otherwise, T moment prior event Y ij The predicted probability of (2) is given by the above P' ijk (T) formula calculation, the formula P' ijk (T) over n iterations, P' ijn (T) is the prior event Y at the moment T ij Predicted probability P' ij The parameter n is the prior event Y ij Of the plurality of characterization events.
7. The event prediction method of claim 6, wherein determining the event rating at which the predicted event occurred based on the one or more first prior events further comprises:
determining a predicted probability for each event class at which one or more first a priori events occurred based on the predicted probability for the one or more first a priori events at the predicted time instant, an
And determining the probability of the event grade of the occurrence of the predicted event according to the prediction probability of each event grade.
8. The event prediction method according to claim 7, wherein the prediction probability of the event level is obtained by the following calculation method:
wherein the event rank X i Has a prediction probability of P " i (T), the prediction probability P " i (T) is in the range of [0,s i ]After m iterations, P ″) im (T) is the event grade X at the moment T i Is expressed as P ″) i The parameter m is shown inIndicating the number of the prior events;
the probability P (T) of the event level of the predicted event occurrence is:
P(T)＝max(P″ i1 ,″p i2 ，p″ i3 ...),
the parameters i1, i2, i3 represent the number of the event classes.
9. The event prediction method of claim 1, wherein the one or more observed data comprises one or more observed events and time information of occurrence of the one or more observed events.
10. The event prediction method of claim 1, further comprising obtaining an event rating at which the predicted event occurs at a particular time.
11. The event prediction method of claim 1, further comprising:
generating one or more policy information based on the level of predicted event occurrence.
12. The event prediction method of claim 1, wherein the event classes include primary, distinct and top three classes.
13. An information processing apparatus for event prediction, comprising a processor, characterized in that the processor is configured to perform the event prediction method according to any one of claims 1-12.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211548149.4A CN115952887A (en) | 2022-12-05 | 2022-12-05 | Event prediction method and information processing apparatus applied to event prediction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211548149.4A CN115952887A (en) | 2022-12-05 | 2022-12-05 | Event prediction method and information processing apparatus applied to event prediction |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115952887A true CN115952887A (en) | 2023-04-11 |
Family
ID=87288468
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211548149.4A Pending CN115952887A (en) | 2022-12-05 | 2022-12-05 | Event prediction method and information processing apparatus applied to event prediction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115952887A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116245139A (en) * | 2023-04-23 | 2023-06-09 | 中国人民解放军国防科技大学 | Training method and device for graph neural network model, event detection method and device |
-
2022
- 2022-12-05 CN CN202211548149.4A patent/CN115952887A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116245139A (en) * | 2023-04-23 | 2023-06-09 | 中国人民解放军国防科技大学 | Training method and device for graph neural network model, event detection method and device |
| CN116245139B (en) * | 2023-04-23 | 2023-07-07 | 中国人民解放军国防科技大学 | Training method and device for graph neural network model, event detection method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Harirchi et al. | Guaranteed model-based fault detection in cyber–physical systems: A model invalidation approach | |
| Gneiting et al. | Comparing density forecasts using threshold-and quantile-weighted scoring rules | |
| CN111737101B (en) | User behavior monitoring method, device, equipment and medium based on big data | |
| DE112017004755T5 (en) | Secondary battery management system with remote parameter estimation | |
| CN111625516B (en) | Method, apparatus, computer device and storage medium for detecting data state | |
| CN103279640B (en) | Satellite borne electronic system FMECA method based on rough set theory and reliability prediction | |
| Zhao et al. | Modeling maxima with autoregressive conditional Fréchet model | |
| CN106485036B (en) | Method for grading asset securitization asset pool based on survival model | |
| CN111080117A (en) | Method and device for constructing equipment risk label, electronic equipment and storage medium | |
| EP3602475A1 (en) | Devices, methods, and systems for determining environmental standard compliance | |
| CN110570544A (en) | method, device, equipment and storage medium for identifying faults of aircraft fuel system | |
| CN107942994A (en) | A kind of satellite temperature control system method for diagnosing faults based on temperature curve feature | |
| CN115841255B (en) | On-site early warning method and system for building engineering based on-line analysis | |
| Sari et al. | Statistical metrics for assessing the quality of wind power scenarios for stochastic unit commitment | |
| CN115952887A (en) | Event prediction method and information processing apparatus applied to event prediction | |
| DE112019003222T5 (en) | State estimating method and apparatus | |
| CN116187552A (en) | Abnormality detection method, computing device, and computer storage medium | |
| CN111680407A (en) | Satellite health assessment method based on Gaussian mixture model | |
| CN105590026A (en) | PCA (Principal Component Analysis) based satellite telemetering regression method | |
| Asefi et al. | Power system anomaly detection and classification utilizing WLS-EKF state estimation and machine learning | |
| CN113988709A (en) | Medium-voltage distribution line fault rate analysis method and device, terminal equipment and medium | |
| CN112199850B (en) | Method for researching system elastic distribution rule based on process characteristics | |
| CN117310500A (en) | Battery state classification model construction method and battery state classification method | |
| Zhang et al. | Extreme wind power forecast error analysis considering its application in day‐ahead reserve capacity planning | |
| CN111783883A (en) | Abnormal data detection method and device |
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 |