CN111814890A - Network live broadcast illegal behavior determination method based on D-S evidence theory - Google Patents

Network live broadcast illegal behavior determination method based on D-S evidence theory Download PDF

Info

Publication number
CN111814890A
CN111814890A CN202010679106.4A CN202010679106A CN111814890A CN 111814890 A CN111814890 A CN 111814890A CN 202010679106 A CN202010679106 A CN 202010679106A CN 111814890 A CN111814890 A CN 111814890A
Authority
CN
China
Prior art keywords
illegal
evidence
probability
behaviors
data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202010679106.4A
Other languages
Chinese (zh)
Other versions
CN111814890B (en
Inventor
王妍
刘德伟
刘迪
田玲玲
谭爱平
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Liaoning University
Original Assignee
Liaoning University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Liaoning University filed Critical Liaoning University
Priority to CN202010679106.4A priority Critical patent/CN111814890B/en
Publication of CN111814890A publication Critical patent/CN111814890A/en
Application granted granted Critical
Publication of CN111814890B publication Critical patent/CN111814890B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/25—Fusion techniques
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/21—Server components or server architectures
    • H04N21/218—Source of audio or video content, e.g. local disk arrays
    • H04N21/2187—Live feed

Landscapes

  • Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Probability & Statistics with Applications (AREA)
  • Databases & Information Systems (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

A method for judging illegal behaviors of live webcasts based on a D-S evidence theory comprises the following steps: 1) multi-source heterogeneous data in a network live broadcast room are input, and a threshold value alpha of illegal behaviors and a threshold value beta of illegal behaviors are obtained according to sample set data; 2) calculating the trust weight of the distributed evidence body by using a Tanimoto coefficient formula; 3) correcting the probability of the violation, the law violation and the serious law violation of the original text/video/audio data by using the evidence body trust degree weight obtained in the step (2); 4) mapping the illegal behavior probability of the corrected text/video/audio data to a unified public space, and performing data fusion by using a D-S evidence theory; 5) and carrying out illegal behavior judgment on the data fusion result. By the method, multi-mode comprehensive evaluation can be performed on the live webcast data, the platform supervision effect is improved, good network ecology is maintained, and a wind-clearing and air-purifying network space is created for vast netizens.

Description

Network live broadcast illegal behavior determination method based on D-S evidence theory
Technical Field
The invention provides a method for judging illegal behaviors of live webcasts based on a D-S evidence theory aiming at characteristic supervision of multisource heterogeneous data of a live webcast platform.
Background
The user scale of the China online live broadcast industry keeps steadily increasing for nearly five years, and the industry development is steady. With the rapid increase of the number of people who live broadcast, live broadcast and delivery become a new industry development trend of the live broadcast industry, and are gradually brought into important strategies by various platforms, and live broadcast electric business is welcomed in the best era. However, while the development is fast, the live webcast platform also becomes a main way to propagate illegal criminal behaviors, which causes a lot of problems: the direct broadcast content is popular, excessive entertainment behaviors such as yellow gambling poison are forbidden frequently, social aversion phenomena such as vague and evil education are built in some places, and sediments are generated, so that great harm is brought to the society.
In the existing live webcast supervision solution, a screenshot is regularly captured in a live webcast background through an internal data interface, and illegal behaviors are analyzed on the screenshot to obtain the probability of the illegal behaviors, so that the illegal behaviors are judged. On one hand, the traditional network live broadcast supervision method only judges illegal behaviors of single modal data, can obtain correct judgment on pictures with clear illegal behavior characteristics, but cannot obtain accurate results on pictures with fuzzy illegal behavior characteristics; on the other hand, when the multi-modal data fusion is used for comprehensively judging illegal behaviors in network live broadcast, certain relation exists among different modes, illegal behavior characteristics among different modes are expressed differently, and the illegal behaviors cannot be fused by using the same weight. The problem to be solved at present is to realize the correlation of multi-mode data to analyze and judge illegal behaviors in a live broadcast room, so that illegal behaviors and the like in network live broadcast are effectively supervised.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides a method for judging illegal behaviors of live webcasts based on a D-S evidence theory. Firstly, inputting multi-source heterogeneous data in a network live broadcast room, and obtaining an illegal behavior threshold value alpha and an illegal behavior threshold value beta according to sample set data; calculating the trust weight of the distributed evidence body by using a Tanimoto coefficient formula; correcting the probability of the illegal, illegal and serious illegal behaviors of the original text/video/audio data by using the calculated evidence body trust degree weight; mapping the illegal behavior probability of the corrected text/video/audio data to a unified public space, and performing data fusion by using a D-S evidence theory; and carrying out illegal behavior judgment on the data fusion result.
In order to achieve the purpose, the invention adopts the technical scheme that:
a method for judging illegal behaviors of live webcasts based on a D-S evidence theory comprises the following steps:
step 1), multi-source heterogeneous data in network live broadcast are input, and a probability threshold value of illegal behaviors is obtained by utilizing a sample set;
step 2), calculating the trust degree weight of the distributed evidence body by using a Tanimoto coefficient formula;
step 3), correcting the probability of the illegal, illegal and serious illegal behaviors of the original text/video/audio data by using the evidence body trust degree weight obtained in the step 2;
step 4), mapping the illegal behavior probability of the obtained text/video/audio data to a unified public space, and performing data fusion on the collected data in the unified public space by using a D-S evidence theory;
and 5) carrying out illegal behavior judgment on the data fusion result.
In the step 1), the specific method comprises the following steps:
1.1) obtaining three types of illegal behaviors, namely no illegal behaviors, possible illegal behaviors and illegal behaviors in the sample set data, and marking the probability of illegal behaviors;
1.2) taking the minimum probability of the illegal behaviors as a threshold alpha and taking the maximum probability of the illegal behaviors as a threshold beta;
1.3) inputting the illegal behavior probability of multi-source heterogeneous data in network live broadcast, if the illegal behavior probability of a single data source is larger than a threshold value alpha, judging that illegal behaviors exist, and if not, continuously judging.
In the step 2), the specific method is as follows:
2.1) utilizing the obtained illegal violation probability matrix to calculate an evidence body correlation matrix S by utilizing a Tanimoto coefficient formula
Figure BDA0002585206610000021
Wherein m isi、mjRepresenting different bodies of evidence, SijRepresents a proof body miAnd a body of evidence mjA correlation coefficient between;
2.2) calculating the confidence weight of the evidence body according to the correlation matrix S, wherein the calculation formula is
Figure BDA0002585206610000022
Wherein S isijRepresents a proof body miAnd a body of evidence mjCoefficient of correlation between, WkRepresents a proof body mkThe confidence weight of.
In the step 3), the specific method is as follows:
3.1) comparing the illegal violation probability with the confidence level ticket w obtained in the step 2iMultiplying to obtain a corrected evidence body probability matrix;
3.2) adding one column to the evidence body probability matrix
Figure BDA0002585206610000023
The values of this column are: 1-sigma m (u)k) Making the probability sum of the same evidence body 1; wherein m (u)k) Is the u th under some evidencekA probability of the individual evidence;
3.3) for s in the corrected probability matrixijAnd performing zero factor repair, setting the zero factor to be 0.1 if the probability representations of other evidence bodies in the original probability table to the focal element are all larger than the threshold value beta, setting the zero factor to be 0.01 if the probability representation of only one evidence body to the focal element is larger than the threshold value beta, otherwise setting the zero factor to be 0.001, and setting the corresponding zero factor to be corresponding to the focal element
Figure BDA0002585206610000024
The value of the column minus the corresponding value.
In the step 4), the specific method is as follows:
4.1) define the data fusion framework U ═ { U ═1,u2,u3,u4}，u1Representing an illegal action, u2Representing an illegal activity, u3Representing severe law violation, u4Representing no illegal acts; for the same identification frame, M ═ M1,m2,m3The basic probability distribution function of evidence body representing that different data sources accord with illegal, illegal and serious illegal behaviors, m1Representing a body of evidence of text data, m2Representing the body of evidence of video data, m3Representing a body of evidence of audio data; for the data fusion framework U ═ U1,u2,u3,u4The basic probability distribution function of is: m isi(uj)＝PijWherein i ═ 1,3],j＝[1,4],ujRepresenting a certain behavior in the fusion framework U, miRepresents the ith evidence body, PijIs represented by miEvidence of the occurrence of u in vivojA probability of a behavior;
4.2) defining a synthesis rule, and conforming the occurrence probability M of illegal, illegal and serious illegal behaviors for the data characteristics of three data sources in M1，m2，m3The occurrence probability of meeting violation, law violation and serious law violation behaviors after fusion is as follows:
Figure BDA0002585206610000031
where k is Σ m1(A)m2(B)m3(C) (A.andgate.B.andgate.C.noteq.empty);
4.3) obtaining a trust interval for the probability of the illegal, illegal and serious illegal behaviors obtained in the step 3 according to the synthesis rule. Evidence uiHas a confidence interval of [0, Bel (u)i)]Wherein
Figure BDA0002585206610000032
i＝[1,3]。
In the step 5), the specific method is as follows:
5.1) arrangement obtained according to step 4Signal interval [0, Bel (u)i)]Judging illegal behaviors, and obtaining the conclusion of illegal, illegal and serious illegal behaviors when the illegal behaviors are larger than a threshold value alpha;
5.2) confidence intervals [0, Bel (u) from step 4i)]And judging illegal behaviors, and obtaining the conclusion of no violation, violation and serious violation when the illegal behaviors are smaller than the threshold alpha.
The beneficial effects created by the invention are as follows:
aiming at the existing problems, the method for judging illegal behaviors of live webcasts based on the D-S evidence theory comprises the steps of firstly inputting multi-source heterogeneous data in a live webcast room, and obtaining an illegal behavior threshold value alpha and an illegal behavior threshold value beta according to sample set data; calculating the trust weight of the distributed evidence body by using a Tanimoto coefficient formula; correcting the probability of the illegal, illegal and serious illegal behaviors of the original text/video/audio data by using the calculated evidence body trust degree weight; mapping the illegal behavior probability of the corrected text/video/audio data to a unified public space, and performing data fusion by using a D-S evidence theory; and carrying out illegal behavior judgment on the data fusion result. The invention can carry out multi-mode comprehensive evaluation on the network live broadcast data, is beneficial to improving the supervision effect of the platform, maintains good network ecology and creates a network space with pure and fresh wind for vast netizens.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
Detailed Description
A network live broadcast illegal behavior judgment method based on a D-S evidence theory comprises the following steps:
1) inputting multi-source heterogeneous data in network live broadcast, and obtaining probability threshold values alpha and beta of illegal behaviors by using a sample set;
(1) obtaining three types of illegal behaviors, namely no illegal behavior, possible illegal behavior and illegal behavior in the sample set data, and marking the probability of illegal behavior;
(2) taking the minimum probability of illegal behaviors as a threshold alpha and the maximum probability of illegal behaviors as a threshold beta;
(3) and inputting the illegal behavior probability of the multi-source heterogeneous data in the network live broadcast, if the illegal behavior probability of the single data source is greater than a threshold value alpha, judging that illegal behaviors exist, and if not, continuously judging.
2) Calculating the confidence weight of the distributed evidence body by using a Tanimoto coefficient formula;
(1) the obtained illegal violation probability is used for solving an evidence body correlation matrix S by using a Tanimoto coefficient formula
Figure BDA0002585206610000041
Wherein m isi、mjRepresenting different bodies of evidence, SijRepresents the evidential entity mi and the evidential entity mjA correlation coefficient between;
(2) calculating the confidence weight of the evidence body according to the correlation matrix S, wherein the calculation formula is
Figure BDA0002585206610000042
Wherein S isijRepresents a proof body miAnd a body of evidence mjCoefficient of correlation between, WkRepresents a proof body mkA confidence weight of;
3) correcting the probability of the illegal, illegal and serious illegal behaviors of the original text/video/audio data by using the evidence body trust degree weight obtained in the step 2;
(1) the illegal violation probability and the trust level ticket w obtained in the step 2 are combinediMultiplying to obtain a corrected evidence body probability matrix;
(2) adding a column to the evidence body probability matrix
Figure BDA0002585206610000043
The values of this column are: 1-sigma m (u)k) Making the probability sum of the same evidence body 1; wherein m (u)k) Is the u th under some evidencekA probability of the individual evidence;
(3) for s in the corrected probability matrixijAnd performing zero factor repair, setting the zero factor to be 0.1 if the probability representations of other evidence bodies in the original probability table to the focal element are all larger than the threshold value beta, setting the zero factor to be 0.01 if the probability representation of only one evidence body to the focal element is larger than the threshold value beta, otherwise setting the zero factor to be 0.001, and setting the corresponding zero factor to be corresponding to the focal element
Figure BDA0002585206610000044
The value of the column minus the corresponding value.
4) Mapping the illegal behavior probability of the obtained text/video/audio data to a unified public space, and performing data fusion on the collected data in the unified public space by using a D-S evidence theory;
(1) defining a data fusion framework U ═ U1,u2,u3,u4}，u1Representative of violations, u2Representing law of violation, u3Representing a serious violation, u4Representing no violation. For the same identification frame, M ═ M1,m2,m3The basic probability distribution function representing that different data sources accord with the behaviors of violation, violation and serious violation, m1Representing text data, m2Representing video data, m3Representing audio data. Data fusion framework U ═ U1,u2,u3,u4The basic probability distribution functions of are: m isi(u1)＝ai,mi(u2)＝bi,mi(u3)＝ci，mi(u4)＝di；
(2) Defining a synthesis rule, and conforming the occurrence probability M of violation, violation and serious violation behaviors for the data characteristics of three data sources in M1，m2，m3The occurrence probability of meeting violation, law violation and serious law violation behaviors after fusion is as follows:
Figure BDA0002585206610000051
where k is Σ m1(A)m2(B)m3(C) (A.andgate.B.andgate.C.noteq.empty);
(3) obtaining a trust interval and evidence u according to the probability of the illegal, illegal and serious illegal behaviors obtained in the step 3 by a synthesis ruleiHas a confidence interval of [0, Bel (u)i)]Wherein
Figure BDA0002585206610000052
i＝[1,3]。
5) And judging illegal behaviors of the data fusion result.
(1) Judging illegal behaviors according to the confidence interval [0, Bel (ui) ] obtained in the step 4, and obtaining the conclusion of illegal, illegal and serious illegal behaviors when the confidence interval is larger than a threshold value alpha;
(2) and (4) judging illegal behaviors according to the confidence interval [0, Bel (ui) ] obtained in the step (4), and obtaining the conclusion of no illegal, illegal and serious illegal behaviors when the confidence interval is smaller than a threshold value alpha.
Example 1:
example (c): supposing that a certain network live broadcast platform utilizes the method to judge illegal network live broadcast behaviors, firstly, the probability of a sample set without illegal behaviors, a sample set with possible illegal behaviors and a sample set with illegal behaviors is obtained according to sample set data, the minimum probability 0.3 of the sample set without illegal behaviors is taken as a threshold beta, the minimum probability 0.6 of the sample set with illegal behaviors is taken as a threshold alpha, and the probability of illegal behaviors obtained by inputting multi-source heterogeneous data in a live broadcast room is shown in a table 1:
table 1: network live broadcast illegal behavior probability table
Figure BDA0002585206610000053
The obtained illegal violation probability is used for solving an evidence body correlation matrix S by using a Tanimoto coefficient formula
Figure BDA0002585206610000054
The resulting matrix S is:
Figure BDA0002585206610000055
calculating the confidence weight of the evidence body according to the correlation matrix S, wherein the calculation formula is
Figure BDA0002585206610000061
The obtained confidence weight W of the evidence body is shown in the table 2:
table 2: evidence body weight table
W1 W2 W3
0.34 0.31 0.35
After the evidence body trust degree weight W is obtained, multiplying the illegal violation probability by the evidence body trust degree weight W, and adding a column to the evidence body probability matrix
Figure BDA0002585206610000067
The values of this column are: 1-sigma m (u)k) The probability sum of the same evidence body is 1, and zero factor correction is performed, and the obtained correction probability is shown in table 3:
table 3: modified illegal behavior probability table
Figure BDA0002585206610000062
After the corrected probability is obtained, D-S data fusion is performed to find K0.007006755
Figure BDA0002585206610000063
Figure BDA0002585206610000064
Figure BDA0002585206610000065
Figure BDA0002585206610000066
The obtained trust interval is:
violation … … … … … … … … … … [0,0.948]
Violation … … … … … … … … … … [0,0.042]
Severe violation … … … … … … … … [0,0.01]
No illegal violation … … … … … … … [0,0]
Confidence interval [0, Bel (u)i)]And integrally judging illegal behaviors to obtain the conclusion of the live broadcast illegal behaviors.

Claims (6)

1. A method for judging illegal behaviors of live webcasts based on a D-S evidence theory is characterized by comprising the following steps:
step 1), multi-source heterogeneous data in network live broadcast are input, and a probability threshold value of illegal behaviors is obtained by utilizing a sample set;
step 2), calculating the trust degree weight of the distributed evidence body by using a Tanimoto coefficient formula;
step 3), correcting the probability of the illegal, illegal and serious illegal behaviors of the original text/video/audio data by using the evidence body trust degree weight obtained in the step 2;
step 4), mapping the illegal behavior probability of the obtained text/video/audio data to a unified public space, and performing data fusion on the collected data in the unified public space by using a D-S evidence theory;
and 5) carrying out illegal behavior judgment on the data fusion result.
2. The method for determining illegal behaviors of live webcasts based on the D-S evidence theory according to claim 1, wherein the specific method in the step 1) is as follows:
1.1) obtaining three types of illegal behaviors, namely no illegal behaviors, possible illegal behaviors and illegal behaviors in the sample set data, and marking the probability of illegal behaviors;
1.2) taking the minimum probability of the illegal behaviors as a threshold alpha and taking the maximum probability of the illegal behaviors as a threshold beta;
1.3) inputting the illegal behavior probability of multi-source heterogeneous data in network live broadcast, if the illegal behavior probability of a single data source is larger than a threshold value alpha, judging that illegal behaviors exist, and if not, continuously judging.
3. The method for determining illegal behaviors of live webcasts based on the D-S evidence theory according to claim 1, wherein in the step 2), the specific method is as follows:
2.1) utilizing the obtained illegal violation probability matrix to calculate an evidence body correlation matrix S by utilizing a Tanimoto coefficient formula
Figure FDA0002585206600000011
Wherein m isi、mjRepresenting different bodies of evidence, SijRepresents a proof body miAnd a body of evidence mjA correlation coefficient between;
2.2) calculating the confidence weight of the evidence body according to the correlation matrix S, wherein the calculation formula is
Figure FDA0002585206600000012
Wherein S isijRepresents a proof body miAnd a body of evidence mjCoefficient of correlation between, WkRepresents a proof body mkThe confidence weight of.
4. The method for determining illegal behaviors of live webcasts based on the D-S evidence theory according to claim 3, wherein in the step 3), the specific method is as follows:
3.1) comparing the illegal violation probability with the confidence level ticket w obtained in the step 2iMultiplying to obtain a corrected evidence body probability matrix;
3.2) adding one column to the evidence body probability matrix
Figure FDA0002585206600000013
The values of this column are: 1-sigma m (u)k) Making the probability sum of the same evidence body 1; wherein m (u)k) Is the u th under some evidencekA probability of the individual evidence;
3.3) for s in the corrected probability matrixijAnd performing zero factor repair, setting the zero factor to be 0.1 if the probability representations of other evidence bodies in the original probability table to the focal element are all larger than the threshold value beta, setting the zero factor to be 0.01 if the probability representation of only one evidence body to the focal element is larger than the threshold value beta, otherwise setting the zero factor to be 0.001, and setting the corresponding zero factor to be corresponding to the focal element
Figure FDA0002585206600000023
The value of the column minus the corresponding value.
5. The method for determining illegal behaviors of live webcasts based on the D-S evidence theory according to claim 1, wherein in the step 4), the specific method is as follows:
4.1) define the data fusion framework U ═ { U ═1,u2,u3,u4}，u1Representing an illegal action, u2Representing an illegal activity, u3Representing severe law violation, u4Representing no illegal acts; for the same identification frame, M ═ M1,m2,m3The basic probability distribution function of evidence body representing that different data sources accord with illegal, illegal and serious illegal behaviors, m1Representing a body of evidence of text data, m2Representative videoBody of evidence of data, m3Representing a body of evidence of audio data; for the data fusion framework U ═ U1,u2,u3,u4The basic probability distribution function of is: m isi(uj)＝PijWherein i ═ 1,3],j＝[1,4],ujRepresenting a certain behavior in the fusion framework U, miRepresents the ith evidence body, PijIs represented by miEvidence of the occurrence of u in vivojA probability of a behavior;
4.2) defining a synthesis rule, and conforming the occurrence probability M of illegal, illegal and serious illegal behaviors for the data characteristics of three data sources in M1，m2，m3The occurrence probability of meeting violation, law violation and serious law violation behaviors after fusion is as follows:
Figure FDA0002585206600000021
where k is Σ m1(A)m2(B)m3(C) (A.andgate.B.andgate.C.noteq.empty);
4.3) obtaining a trust interval for the probability of the illegal, illegal and serious illegal behaviors obtained in the step 3 according to the synthesis rule. Evidence uiHas a confidence interval of [0, Bel (u)i)]Wherein
Figure FDA0002585206600000022
6. The method for determining illegal behaviors of live webcasts based on the D-S evidence theory according to claim 5, wherein in the step 5), the specific method is as follows:
5.1) confidence intervals [0, Bel (u) from step 4i)]Judging illegal behaviors, and obtaining the conclusion of illegal, illegal and serious illegal behaviors when the illegal behaviors are larger than a threshold value alpha;
5.2) confidence intervals [0, Bel (u) from step 4i)]And judging illegal behaviors, and obtaining the conclusion of no violation, violation and serious violation when the illegal behaviors are smaller than the threshold alpha.
CN202010679106.4A 2020-07-15 2020-07-15 Method for judging illegal and illegal behaviors of live webcast based on D-S evidence theory Active CN111814890B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010679106.4A CN111814890B (en) 2020-07-15 2020-07-15 Method for judging illegal and illegal behaviors of live webcast based on D-S evidence theory

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010679106.4A CN111814890B (en) 2020-07-15 2020-07-15 Method for judging illegal and illegal behaviors of live webcast based on D-S evidence theory

Publications (2)

Publication Number Publication Date
CN111814890A true CN111814890A (en) 2020-10-23
CN111814890B CN111814890B (en) 2024-09-10

Family

ID=72864795

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010679106.4A Active CN111814890B (en) 2020-07-15 2020-07-15 Method for judging illegal and illegal behaviors of live webcast based on D-S evidence theory

Country Status (1)

Country Link
CN (1) CN111814890B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112583804A (en) * 2020-12-05 2021-03-30 星极实业(深圳)有限公司 Monitoring management system capable of tracking and evidence obtaining of network illegal behaviors in real time
CN113887446A (en) * 2021-10-08 2022-01-04 黑龙江雨谷科技有限公司 Audio and video combined pedestrian accidental falling monitoring method based on neural network

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100010776A1 (en) * 2008-07-10 2010-01-14 Indranil Saha Probabilistic modeling of collaborative monitoring of policy violations
CN106154182A (en) * 2016-08-26 2016-11-23 上海电力学院 A kind of based on the lithium battery method for diagnosing faults improving D S evidence theory
CN106250935A (en) * 2016-08-15 2016-12-21 广东石油化工学院 The rotary machinery fault diagnosis method that genetic programming and weighted evidence theory merge
CN110119713A (en) * 2019-05-14 2019-08-13 云南电网有限责任公司电力科学研究院 Mechanical Failure of HV Circuit Breaker diagnostic method based on D-S evidence theory
CN110533091A (en) * 2019-08-22 2019-12-03 贵州大学 A kind of more evident information fusion methods for improving DS evidence theory
CN111881416A (en) * 2020-07-13 2020-11-03 雷震烁 Method for processing conflict evidence based on DS evidence theory

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100010776A1 (en) * 2008-07-10 2010-01-14 Indranil Saha Probabilistic modeling of collaborative monitoring of policy violations
CN106250935A (en) * 2016-08-15 2016-12-21 广东石油化工学院 The rotary machinery fault diagnosis method that genetic programming and weighted evidence theory merge
CN106154182A (en) * 2016-08-26 2016-11-23 上海电力学院 A kind of based on the lithium battery method for diagnosing faults improving D S evidence theory
CN110119713A (en) * 2019-05-14 2019-08-13 云南电网有限责任公司电力科学研究院 Mechanical Failure of HV Circuit Breaker diagnostic method based on D-S evidence theory
CN110533091A (en) * 2019-08-22 2019-12-03 贵州大学 A kind of more evident information fusion methods for improving DS evidence theory
CN111881416A (en) * 2020-07-13 2020-11-03 雷震烁 Method for processing conflict evidence based on DS evidence theory

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
陈圣磊 等: "D-S证据理论在审计证据融合中的应用研究", 计算机工程与应用, pages 27 - 29 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112583804A (en) * 2020-12-05 2021-03-30 星极实业(深圳)有限公司 Monitoring management system capable of tracking and evidence obtaining of network illegal behaviors in real time
CN113887446A (en) * 2021-10-08 2022-01-04 黑龙江雨谷科技有限公司 Audio and video combined pedestrian accidental falling monitoring method based on neural network

Also Published As

Publication number Publication date
CN111814890B (en) 2024-09-10

Similar Documents

Publication Publication Date Title
CN107403092B (en) Mobile phone APP privacy risk quantitative evaluation method
CN110222058A (en) Multi-source data based on FP-growth is associated with privacy leakage risk evaluating system
CN109491915B (en) Data processing method and device, medium and computing equipment
CN111814890A (en) Network live broadcast illegal behavior determination method based on D-S evidence theory
CN113283795B (en) Data processing method and device based on two-classification model, medium and equipment
CN111582394B (en) Group assessment method, device, equipment and medium
CN110958472A (en) Video click rate rating prediction method and device, electronic equipment and storage medium
WO2018036402A1 (en) Method and device for determining key variable in model
CN113435713B (en) Risk map compiling method and system based on GIS technology and two-model fusion
CN110650043A (en) Key business system identification and risk assessment method for business process
Zhu et al. Gaussian mixture model based prediction method of movie rating
CN113723747A (en) Analysis report generation method, electronic device and readable storage medium
JP5937529B2 (en) Time-series data component decomposition apparatus, method, program, and recording medium
CN114648688B (en) High-speed rail line landscape grade assessment method, system, equipment and readable storage medium
CN115796600A (en) Public opinion risk early warning method, system, medium and electronic equipment
CN111667339B (en) Defamation malicious user detection method based on improved recurrent neural network
CN115471036A (en) Group emotion analysis method, storage medium and device for hotspot events
Ballerini et al. Fisher's Noncentral Hypergeometric Distribution for Population Size Estimation
TWI722327B (en) Audio-visual content and user interaction sequence analysis system and method
JP2011186762A (en) Document evaluation imparting method, program, and device
CN111539834A (en) Pre-check method and device
CN113128217B (en) Public opinion disposition decision-making method based on network twinning space
CN118333585B (en) Application range review processing method, device, equipment and medium for management system authentication
CN114398638B (en) Network black product identification method, device, equipment and medium
CN110879723B (en) Objective evaluation method and device for software service value based on Pareto optimal set

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant