CN105956982A - Method of predicting act of terror based on background change - Google Patents
Method of predicting act of terror based on background change Download PDFInfo
- Publication number
- CN105956982A CN105956982A CN201610300731.7A CN201610300731A CN105956982A CN 105956982 A CN105956982 A CN 105956982A CN 201610300731 A CN201610300731 A CN 201610300731A CN 105956982 A CN105956982 A CN 105956982A
- Authority
- CN
- China
- Prior art keywords
- behavior
- change
- background
- change table
- prediction
- 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
- 230000008859 change Effects 0.000 title claims abstract description 93
- 238000000034 method Methods 0.000 title claims abstract description 37
- 230000006399 behavior Effects 0.000 claims description 81
- 230000008520 organization Effects 0.000 claims description 14
- 239000000284 extract Substances 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 230000007613 environmental effect Effects 0.000 claims description 2
- 238000002372 labelling Methods 0.000 claims description 2
- 239000000203 mixture Substances 0.000 claims 1
- 238000013398 bayesian method Methods 0.000 abstract 3
- 230000002045 lasting effect Effects 0.000 abstract 1
- 238000004422 calculation algorithm Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 4
- 239000000523 sample Substances 0.000 description 4
- 230000003542 behavioural effect Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000007418 data mining Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000013277 forecasting method Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000005923 long-lasting effect Effects 0.000 description 2
- ABEXEQSGABRUHS-UHFFFAOYSA-N 16-methylheptadecyl 16-methylheptadecanoate Chemical compound CC(C)CCCCCCCCCCCCCCCOC(=O)CCCCCCCCCCCCCCC(C)C ABEXEQSGABRUHS-UHFFFAOYSA-N 0.000 description 1
- 241000764238 Isis Species 0.000 description 1
- 238000009412 basement excavation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000034994 death Effects 0.000 description 1
- 231100000517 death Toxicity 0.000 description 1
- 238000005417 image-selected in vivo spectroscopy Methods 0.000 description 1
- 238000012739 integrated shape imaging system Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000003449 preventive effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 235000015170 shellfish Nutrition 0.000 description 1
- 230000035939 shock Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- G06Q50/26—Government or public services
- G06Q50/265—Personal security, identity or safety
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Computer Security & Cryptography (AREA)
- Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Educational Administration (AREA)
- General Health & Medical Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method of predicting an act of terror based on background change, mainly comprising the following steps: generating a change table, and predicting acts using a Bayesian method; during generation of an h- change table, combining the background change at the current time and the act change after h cycles to form a record in the change table; during prediction, under the premise of inputting a background change vector, using the Bayesian method to calculate a classification result of maximum probability from the change table so as to predict an act after h cycles; building a weighted Bayesian model in view of the fact that background change may have a lasting impact on the organizational act in time series; and using the model to calculate the probability of each act in different change tables under the condition that the time lag is 1 to H respectively, and finally, weighting the probability of all kinds of acts. Through the method, an act of terror can be predicted according to any background change. By using the Bayesian method, the problem of high-dimensional small sample classification can be solved quickly and efficiently, and the precision of prediction is improved.
Description
Technical field
The invention belongs to the organizational behavior Predicting Technique in data mining, be specifically related to a kind of on the change table improved, use
Bayes method predicts terroristic algorithm.
Background technology
The act of terrorism refers to that unarmed personnel are in an organized way used force or try threat of violence by implementer, by by certain
Object is placed among terror, reaches the purpose on religion, politics or ideology.The U.S. " 9 11 " event shocks
The whole world so that terroristic organization becomes the focus in the whole world, becomes one of factor affecting world peace.Recently, Islam
The state (ISIS) activity in the Middle East grows in intensity.This is organized in and starts vindictive bombing raid at Paris, FRA in November, 2014
Event, causes the injures and deaths of substantial amounts of innocent people, and the attack of terrorism has become to be primarily upon target again.How to utilize existing
The behavior of information prediction terroristic organization, becomes an important problem.
Act of terrorism prediction is typical case's application of knowledge excavation, it utilize the correlation technique of data mining and machine learning to the past,
The terroristic situation that terroristic organization's planning now is implemented carries out the statistical analysis of science, then predicts its development trend.Probably
Fear prediction is not to confirm in the past, is not explanation reality, but Probe into Future, it was predicted that the development trend of terrorist's organization.
Its object is to as taking effective preventive measure to provide decision support.The reason that the attack of terrorism occurs comprises politics, warp
The factor of the aspects such as Ji and culture, various factors weave in, make act of terrorism prediction seem intricate.To terrified row
The information such as time, place and influence degree that terrorist incident occurs can not be only considered for the research of prediction, and should be with all kinds of
Based on section's (computer data digging technology and sociology, criminology etc.), consider include terroristic organization politics,
Culture, economic dispatch contextual factor, and provide more effectively support by these data being analyzed the decision-making into authorities.Therefore,
The relation between background knowledge and its behavior of terroristic organization of analyzing becomes the focus of research.
At present, predict that the basic ideas of terroristic Forecasting Methodology are essentially identical, i.e. according to background tissue based on background knowledge
And the relation between behavior, builds the behavior prediction model of tissue.Specifically, these methods utilize between background vector
The behavior vector that similarity prediction is corresponding, then the behavior vector of prediction is carried out certain calculating, obtain each behavior in behavior vector
Probability, be then given according to the probability of happening of each behavior and predict the outcome.But, in reality, tissue typically has counter detecing
Looking into ability, therefore the attribute such as time, place and intensity of performance that its terrorist activity occurs can change.Existing most models
This change not accounting for tissue and the Behavioral change thus caused.And the change that take into account between background and behavior contacts
Model, although attempt by learning organization change behavior condition, construct the rule change between background and behavior, but
There is also many shortcomings.Such as, model can only be according to background variation prediction behavior present in change table, and predictablity rate
Relatively low and time complexity is high.
The present invention gives a kind of act of terrorism prediction algorithm based on change table and bayes method, this algorithm both can be effective
Extract terroristic background knowledge subspace, can ensure that again the effectiveness utilizing the background subspace extracted to be predicted.
Summary of the invention
The present invention may cause this thought of change of behavior according to the background change of tissue, utilizes the background knowledge of terrorist
Predict its behavior.In order to predict organizational behavior under arbitrary background changes, for data set higher-dimension small sample feature,
A kind of act of terrorism prediction algorithm based on bayes method with change table is proposed.The change table improved stores background and behavior
Existing situation of change, makes bayes method can the most therefrom collect relevant change information.For new background change,
Bayes method carries out comprehensive descision according to each element in input background change vector to the influence degree of different classification results,
It is thus possible to realize the purpose of prediction behavior under arbitrary background changes.In addition, for higher-dimension Small Sample Database, shellfish
This method of leaf can be predicted fast and effectively.Finally, it is contemplated that changing the impact of behavior of background is not instantaneous, but
Within cycle regular hour, behavior can be produced long lasting effect.Therefore, the method considers under different time decalage comprehensive
Prediction organizational behavior.
The present invention utilizes the background knowledge of terroristic organization to extract the Forecasting Methodology of subspace, comprises the following steps:
(1) pretreatment of raw data set；
Initial data is made up of the essential information of terroristic organization, background knowledge and behavioral activity, extracts background knowledge and behavior,
The vector being marked as (CS (g), AS (g)) is right.Its CS (g)=(C1,C2,…,CM) represent the background attribute in data, AS (g)
=(A1,A2,…,AN) represent the behavior property related in data.In order to obtain the background subspace of different behavior, need logarithm
According to doing pretreatment, form (CS (g), Aj) N number of Sub Data Set.If terroristic organization, Tg represent group in data set to make g represent
Knit behavior AjSub Data Set, then Tg=(x1,x2,...,xm)T, xi=(ci1,ci2,...,cin,aij), wherein ciBelong to background attribute collection CS (g),
ajBelong to behavior property collection AS (g).
(2) for each behavior to be predicted in AS (g), the change table improved is generated；
Change table is that the behavior in time series of a kind of record organization, with the data store organisation of its background variation relation, represents the back of the body
The change of the scape impact on Behavioral change.For integer h >=1, h-changes table CTh(g,AjIn), the record of time period i is belonged to by background
Property change from time period i-h to i-h+1 and behavior property change from i to i+1 produce.Comprise background as shown in table 1
With behavior AjInitial data after treatment, generate behavior A as shown in table 2jH-change table CTh(g,Aj).Described
Change table includes 1 step change table and multistep change table.
In table 2, PAjRepresent the behavior in last cycle, LAJRepresent the behavior of current time.
(3) utilize bayes method, the change table generated calculates and predicts the outcome；
(4) use multistep weighting Bayesian model, the change table of different step-lengths is predicted respectively, and provides integrated forecasting
Result:
1) according to step number H, the change table CT of 1 to H step is builth(g,Aj), h=1 ..., H；；
2) table CT is changed with h-h(g,AjThe delta data of the every a line in) uses bayes method prediction, obtains the behavior of prediction
Sequence PYh；
3) according to formula rh=Cov (PYh,Yh)/(D(PYh)*D(Yh))1/2(wherein Cov (.) is covariance, and D (.) is variance) counts
Calculate prediction behavior sequence PYhWith real behavior sequence YhCorrelation coefficient rh；
4) judge whether to terminate？If it is not, jump to 2), calculate the correlation coefficient of next step-length；Otherwise, carry out
Next step.
5) formula w is usedh=| rh|/Σ|ri| the correlation coefficient r that will obtain1,r2,...,rHNormalization, obtains predicting weight coefficient
w1,w2,...,wH；Σ|ri| represent and correlation coefficient is sued for peace, i=1,2 ... H.
6) for every kind of behavior, each step change table uses Bayes's classification prediction, calculates its object function maxi{Σh
(wh*Ph(Ci|X)/ΣkPh(Ck| X)) }, select to make the maximized behavior of result value as classification results.Wherein, CiRepresent every
The behavior of kind；H=1 ..., H, represent the step number of change table.The relational expression of object function represents respectively on 1 to H step change table,
Calculating behavior CiWeighting proportion in the probability of happening of all behaviors.
Forecasting Methodology in the present invention is divided into two steps: first generate corresponding change table according to the specific act of terrorism,
Then a situation arises to predict this behavior in the change table generated.
In the present invention during the generation of change table affect decalage h represent background attribute change can h time cycle it
Afterwards behavior property change is produced impact, and then uses the act of terrorism after bayes method prediction h step.And multistep pattra leaves
This neutralizes prediction, considers the prediction case under the influence of 1 to h step decalage, makes final predicting the outcome.
The present invention mainly has a beneficial effect in terms of following two:
(1) in terms of prediction algorithm
The feature of higher-dimension small sample present in raw data set, causes the less effective of general forecast method.And Bayes side
Method solves this problem and has natural advantage, it is possible to make a prediction fast and effectively.Finally, it is contemplated that the change of background is right
The impact of behavior is not instantaneous, but within cycle regular hour, behavior can be produced long lasting effect.Therefore, algorithm
Consider integrated forecasting organizational behavior under different time decalage.For every kind of behavior, calculate and use on each step change table
Bayes's classification prediction obtains the probability of the behavior proportion sum in all classification results probability, selects to make result value
The behavior of bigization is as classification results.
(2) in terms of change table generation
The change table improved stores background and the existing situation of change of behavior, makes bayes method can the most therefrom collect phase
The change information closed.For new background change, bayes method uses certain method to calibrate, and becomes according to input background
Change each element in vector and the influence degree of different classification results is carried out comprehensive descision such that it is able to change in arbitrary background
The purpose of lower realization prediction behavior.And the decalage h that affects during changing the generation of table represents that the change of background attribute can be at h
After the individual time cycle, behavior property change is produced impact, and then uses the act of terrorism after bayes method prediction h step.
Accompanying drawing explanation
Fig. 1 is the flow chart that in the present invention, the training module of multistep weighting Bayesian forecasting method calculates weight.
Fig. 2 is the flow chart using multistep weighting Bayesian forecasting method prediction module prediction organizational behavior in the present invention.
Detailed description of the invention
Below as a example by the background data subset shown in table 3, the multistep in conjunction with Fig. 1, Fig. 2 weights in bayes predictive model
Weight calculation and behavior prediction flow process, describe the algorithm flow of the present invention in detail.
Table 3 is provided with eight fields altogether, is respectively labeled as Time, C1, C2, C3, C4, C5, C6 and A, wherein Time
It is labeled as record time series in table, C1, C2, C3, C4, C5, C6}=CS (g) belong to background knowledge attribute,
A belongs to behavior property.Whether purpose i.e. utilizes the generation of background knowledge prediction prediction act of terrorism A shown in table 3.
Specifically comprise the following steps that
(1) utilize the data genaration 1 step change table of table 3, obtain change table as shown in table 4.
(2) according to bayes method, change table is carried out statistical classification, obtains classification results as shown in table 5.
In table 5, use Bayes's classification need to calculate P (LA=0 | X) and P (LA=1 | X), the then size of class probability, will
The class labelling of greater probability predicts the outcome as classification.Wherein, according to Bayes theorem, compare P (LA=0 | X) and
The size of P (LA=1 | X) is equivalent to compare the size of P (X | LA=0) P (LA=0) and P (X | LA=1) P (LA=1).Wherein,
P (X | LA=0) P (LA=0) and P (X | LA=1) P (LA=1) is illustrated respectively under environmental change vector X, and behavior is probability and the row of 0
For for 1 probability.
Use the result of 1 step change table prediction, be the behavior after prediction 1 time cycle of terroristic organization.For example, it is assumed that
Current background is changed to: (0,0), (1,1), (0,1), (0,0), (1,0), (1,0), 0}, then the behavior of prediction next time cycle is 1,
I.e. there is such terrorist activity.
(3) multistep weighting Bayes's integrated forecasting is used, it is assumed that given step number is 2.
First, method as shown in table 2, set up 1 step and 2 step change tables, result such as table 4, shown in table 6.
2. secondly, according to bayes method, 1 step and 2 step change tables are carried out statistical classification, obtain result such as table 5, table 7
Shown in.
The most then, use the behavior sequence that obtains of prediction respectively with real behavior sequence, calculate the correlation coefficient of each step,
Completing normalization, as the weight of this step change table prediction, result is as shown in table 8.
When 4. predicting, it is assumed that current background change vector X is: { (0,0), (1,1), (0,1), (0,0), (1,0), (1,0), 0}, exist respectively
Step change table often calculates P (LA=0 | X) and P (LA=1 | X), uses P (X | LA=0) P (LA=0) and P (X | LA=1) P (LA=1) equally
Being replaced, result of calculation is as shown in table 9.
As LA=0, Σh(wh*Ph(Ci|X)/ΣkPh(Ck| X))=0.027*0.5+0.0610*0.5=0.044；
As LA=1, Σh(wh*Ph(Ci|X)/ΣkPh(Ck| X))=0.9730*0.5+0.9390*0.5=0.956；
Because 0.044 is less than 0.956, it was predicted that result is LA=1, and terrorist activity i.e. occurs.
The a series of detailed description of those listed above is only for illustrating of the feasibility embodiment of the present invention, it
And be not used to limit the scope of the invention, all without departing from the skill of the present invention equivalent implementations made of spirit or change
Should be included within the scope of the present invention.
Claims (4)
1. one kind utilizes the terroristic method of background variation prediction, it is characterised in that utilize the change between background and behavior
Change contact prediction organizational behavior, comprise the steps:
(1) pretreatment of raw data set；Described initial data is by the essential information of terroristic organization, background knowledge and behavior
Movable composition, extracts background knowledge and behavior, and the vector being marked as (CS (g), AS (g)) is right；Wherein CS (g)=
(C1,C2,…,CM) represent the background attribute in data, AS (g)=(A1,A2,…,AN) represent the behavior property related in data；
In order to obtain the background subspace of different behavior, need initial data is done pretreatment, form (CS (g), Aj) N number of son
Data set；
(2) for each behavior to be predicted in AS (g), the change table improved is generated；The change table of described improvement is a kind of
Record organization behavior in time series, with the data store organisation of its background variation relation, represents that behavior is become by the change of background
The impact changed, its storage content includes background and the existing situation of change of behavior；Described change table includes that 1 step change table is with many
Step change table.
(3) utilize bayes method, 1 step generated or multistep change table calculate and predicts the outcome；
(4) use multistep weighting Bayesian model, the change table that 1 step to H walks is predicted respectively, and provides integrated forecasting
Result.
One the most according to claim 1 utilizes the terroristic method of background variation prediction, it is characterised in that institute
The method of generation change table in step (2) of stating includes: change table CT for integer h > 1, h-h(g,AjThe note of time period i in)
Record the change from i to i+1 by background attribute change from time period i-h to i-h+1 and behavior property to produce.
One the most according to claim 1 utilizes the terroristic method of background variation prediction, it is characterised in that institute
The realization stating step (3) includes: change table carries out statistical classification, calculates P (LAj| X), then according to behavior probability
Size, using the class labelling of greater probability as classification results；Wherein X represents environmental change vector, and LA represents current time
Behavior.
One the most according to claim 1 utilizes the terroristic method of background variation prediction, it is characterised in that step
Suddenly the realization of (4) comprises the steps:
1) according to step number H, the change table CT of 1 to H step is builth(g,Aj), h=1 ..., H；
2) h-is used to change table CTh(g,AjThe delta data of the every a line in) uses bayes method prediction, obtains the row of prediction
For sequence PYh；
3) according to rh=Cov (PYh,Yh)/(D(PYh)*D(Yh))1/2；Wherein Cov (.) is covariance, and D (.) is variance, calculates pre-
Survey behavior sequence PYhWith real behavior sequence YhCorrelation coefficient rh；
4) judge whether to terminate, if it has not ended, then jump to 2), calculate the correlation coefficient of next step-length；Otherwise,
Carry out next step.
5) formula w is usedh=| rh|/Σ|ri| the correlation coefficient r that will obtain1,r2,...,rHNormalization, obtains predicting weight coefficient
w1,w2,...,wH；
6) for every kind of behavior, each step change table uses Bayes's classification prediction, calculates its object function maxi{Σh
(wh*Ph(Ci|X)/ΣkPh(Ck| X)) }, select to make the maximized behavior of result value as classification results；Wherein, CiRepresent every
The behavior of kind；H=1 ..., H, represent the step number of change table；The relational expression of object function represents respectively on 1 to H step change table,
Calculating behavior CiWeighting proportion in the probability of happening of all behaviors.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610300731.7A CN105956982A (en) | 2016-05-04 | 2016-05-04 | Method of predicting act of terror based on background change |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610300731.7A CN105956982A (en) | 2016-05-04 | 2016-05-04 | Method of predicting act of terror based on background change |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105956982A true CN105956982A (en) | 2016-09-21 |
Family
ID=56914087
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610300731.7A Pending CN105956982A (en) | 2016-05-04 | 2016-05-04 | Method of predicting act of terror based on background change |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105956982A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106776884A (en) * | 2016-11-30 | 2017-05-31 | 江苏大学 | A kind of act of terrorism Forecasting Methodology that multi-categorizer is combined based on multi-tag |
| CN108776817A (en) * | 2018-06-04 | 2018-11-09 | 孟玺 | The type prediction method and system of the attack of terrorism |
| CN109685321A (en) * | 2018-11-26 | 2019-04-26 | 山东师范大学 | Event risk method for early warning, electronic equipment and medium based on data mining |
| CN110009022A (en) * | 2019-03-26 | 2019-07-12 | 第四范式（北京）技术有限公司 | Prediction technique, device and the calculating equipment of drug addict's information |
-
2016
- 2016-05-04 CN CN201610300731.7A patent/CN105956982A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106776884A (en) * | 2016-11-30 | 2017-05-31 | 江苏大学 | A kind of act of terrorism Forecasting Methodology that multi-categorizer is combined based on multi-tag |
| CN106776884B (en) * | 2016-11-30 | 2021-04-20 | 江苏大学 | Terrorism prediction method based on multi-label combination and multi-classifier |
| CN108776817A (en) * | 2018-06-04 | 2018-11-09 | 孟玺 | The type prediction method and system of the attack of terrorism |
| CN109685321A (en) * | 2018-11-26 | 2019-04-26 | 山东师范大学 | Event risk method for early warning, electronic equipment and medium based on data mining |
| CN110009022A (en) * | 2019-03-26 | 2019-07-12 | 第四范式（北京）技术有限公司 | Prediction technique, device and the calculating equipment of drug addict's information |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2022135265A1 (en) | Failure warning and analysis method for reservoir dispatching rules under effects of climate change | |
| Zhai et al. | Doro: Distributional and outlier robust optimization | |
| Li et al. | Energy data generation with wasserstein deep convolutional generative adversarial networks | |
| CN105956982A (en) | Method of predicting act of terror based on background change | |
| CN106776884A (en) | A kind of act of terrorism Forecasting Methodology that multi-categorizer is combined based on multi-tag | |
| Finke et al. | Boosting mono-jet searches with model-agnostic machine learning | |
| CN104899430A (en) | Multisource example transfer learning-based terror act prediction method | |
| Leigh et al. | $\nu $-flows: Conditional neutrino regression | |
| Leka et al. | Solar flare forecasting: Present methods and challenges | |
| Depeweg et al. | Sensitivity analysis for predictive uncertainty in Bayesian neural networks | |
| Ting et al. | Mining of stock data: intra-and inter-stock pattern associative classification | |
| Doostan et al. | Predicting lightning-related outages in power distribution systems: A statistical approach | |
| Riley et al. | Surrogate forward models for population inference on compact binary mergers | |
| Lima Junior et al. | An experimental study of fitness function and time series forecasting using artificial neural networks | |
| Graff et al. | Modeling the swift bat trigger algorithm with machine learning | |
| Hossain et al. | Predicting the success of suicide terrorist attacks using different machine learning algorithms | |
| Alshammari et al. | Identifying Flare-indicative Photospheric Magnetic Field Parameters from Multivariate Time-series Data of Solar Active Regions | |
| Li et al. | Classify the Higgs decays with the PFN and ParticleNet at electron–positron colliders | |
| Hou | Time‐varying relationship between inflation and inflation uncertainty | |
| CN103065047A (en) | Terrorism behavior prediction method based on terrorist organization background knowledge subspace | |
| Pasquato et al. | Interpretable machine learning for finding intermediate-mass black holes | |
| Gustian et al. | Comparison data mining based on optimization algorithms in receiving electricity subsidies | |
| Oltean et al. | An econophysical approach of polynomial distribution applied to income and expenditure | |
| Chierichini et al. | CME Arrival Modeling with Machine Learning | |
| Wang et al. | Research on quantitative investment strategies based on deep learning algorithms in the context of the need for information management |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160921 |
|
| RJ01 | Rejection of invention patent application after publication |