CN109523088A - The abnormal behaviour forecasting system of forced quarantine addict received treatment based on big data - Google Patents
The abnormal behaviour forecasting system of forced quarantine addict received treatment based on big data Download PDFInfo
- Publication number
- CN109523088A CN109523088A CN201811451746.9A CN201811451746A CN109523088A CN 109523088 A CN109523088 A CN 109523088A CN 201811451746 A CN201811451746 A CN 201811451746A CN 109523088 A CN109523088 A CN 109523088A
- Authority
- CN
- China
- Prior art keywords
- predicted
- person
- behavior
- abnormal
- abnormal behavior
- 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
- 206010000117 Abnormal behaviour Diseases 0.000 title claims abstract description 250
- 238000013500 data storage Methods 0.000 claims abstract description 12
- 238000007405 data analysis Methods 0.000 claims abstract description 8
- 230000004044 response Effects 0.000 claims abstract description 7
- 230000006399 behavior Effects 0.000 claims description 92
- 239000003814 drug Substances 0.000 claims description 73
- 229940079593 drug Drugs 0.000 claims description 73
- 238000002955 isolation Methods 0.000 claims description 54
- 238000013145 classification model Methods 0.000 claims description 34
- 238000007781 pre-processing Methods 0.000 claims description 25
- 238000012549 training Methods 0.000 claims description 21
- 208000011117 substance-related disease Diseases 0.000 claims description 18
- 206010013663 drug dependence Diseases 0.000 claims description 15
- 238000007726 management method Methods 0.000 claims description 15
- 238000012360 testing method Methods 0.000 claims description 14
- 230000001066 destructive effect Effects 0.000 claims description 12
- 238000011156 evaluation Methods 0.000 claims description 12
- 238000005070 sampling Methods 0.000 claims description 12
- 238000004458 analytical method Methods 0.000 claims description 11
- 206010010144 Completed suicide Diseases 0.000 claims description 10
- 238000000034 method Methods 0.000 claims description 6
- 206010065604 Suicidal behaviour Diseases 0.000 claims description 5
- 208000019901 Anxiety disease Diseases 0.000 claims description 3
- 206010013654 Drug abuse Diseases 0.000 claims description 3
- 230000036506 anxiety Effects 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 claims description 3
- 230000029052 metamorphosis Effects 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 claims description 3
- 230000002159 abnormal effect Effects 0.000 claims description 2
- 238000001784 detoxification Methods 0.000 claims description 2
- 230000003533 narcotic effect Effects 0.000 abstract 2
- 241001269238 Data Species 0.000 abstract 1
- 230000003542 behavioural effect Effects 0.000 abstract 1
- 210000000352 storage cell Anatomy 0.000 abstract 1
- 230000007547 defect Effects 0.000 description 5
- 241001622623 Coeliadinae Species 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- 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 the abnormal behaviour forecasting systems of the forced quarantine addict received treatment based on big data, comprising: data acquisition unit, data storage cell, data analysis unit, data application unit；The data acquisition unit acquires the data information of forced quarantine addict received treatment from the operation system and security system of forced quarantine narcotic house；The data analysis unit carries out model learning according to the data information of known people, obtains learning model, and predict using the abnormal behaviour that learning model treats prognosticator, obtains prediction result；The data application unit is according to prediction result as relevant response.The present invention is based on big datas to construct behavioural characteristic integrated study model, the abnormal behaviour of forced quarantine addict received treatment is precisely predicted and alarmed to realize, effectively the abnormal behaviour of forced quarantine addict received treatment is taken precautions against, safety precaution and the personnel's supervision for improving forced quarantine narcotic house are horizontal.
Description
Technical Field
The invention relates to the technical field of big data-based prediction analysis, in particular to a big data-based abnormal behavior prediction system for forcibly isolating drug addicts.
Background
The forced isolation of the drug-dropping personnel is easy to have unknown abnormal behaviors which cannot be judged, such as escape, suicide, violence, damage and the like, and the unknown abnormal behaviors which cannot be judged bring unprecedented challenges to the safety and stability of the drug-dropping system and personnel supervision and become a problem which must be solved for the informatization construction of the drug-dropping system.
At present, the abnormal behavior assessment of the forced isolation drug-dropping personnel is mainly carried out by means of manual experience assessment, psychological test scale assessment and single classification model assessment, and the three methods have the following problems: firstly, the abnormal behavior cannot be comprehensively and accurately predicted due to excessive dependence on manual evaluation; secondly, the classification of the abnormal behavior is not subdivided, which is not beneficial for policemen to adopt targeted measures to prevent so as to correspond to the occurrence of different abnormal behaviors; thirdly, classification of abnormal behaviors depends heavily on manually designed indexes, and prediction performance is poor. Therefore, the existing abnormal behavior evaluation mode for the forced isolation drug addict is difficult to achieve ideal classification precision and stability.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention provides the abnormal behavior prediction system of the forced isolation drug-relief personnel based on the big data, and the behavior characteristic integrated learning model of the forced isolation drug-relief personnel based on the big data is constructed by combining the safety and stability and the actual personnel supervision requirements of the forced isolation drug-relief personnel, so that the abnormal behavior of the forced isolation drug-relief personnel is accurately predicted and alarmed, the abnormal behavior of the forced isolation drug-relief personnel is effectively prevented, and the safety precaution and personnel supervision level of the forced isolation drug-relief personnel is improved.
In order to achieve the purpose, the invention adopts the following technical scheme that:
the abnormal behavior prediction system of the person who forces to keep apart drug addiction based on big data, characterized by that, the said system includes: the device comprises a data acquisition unit (1), a data storage unit (2), a data analysis unit (3) and a data application unit (4);
the data acquisition unit (1) is connected with a business system and a security system of the forced isolation drug rehabilitation institute, and acquires data information of the forced isolation drug rehabilitation personnel from the business system and the security system of the forced isolation drug rehabilitation institute; the data acquisition unit (1) stores the acquired data information of the forced isolation drug rehabilitation personnel to the data storage unit (2);
the data analysis (3) unit comprises a data preprocessing module (31), an analysis model module (32) and a prediction early warning module (33); wherein,
the data preprocessing module (31) acquires data information of the person who forcibly isolates the drug to be stopped from the data storage unit (2), and constructs characteristic attribute items of the person who forcibly isolates the drug to be stopped according to the data information of the person who forcibly isolates the drug to be stopped from the drug to be stopped; the data preprocessing module (31) carries out numerical preprocessing on the characteristic attribute items of the forced isolation drug-dropping personnel to obtain tuples of the forced isolation drug-dropping personnel; the data preprocessing module (31) also marks tuples of the forced isolation drug addicts, namely known people, which are judged to have abnormal behaviors, so as to obtain the marked tuples of the known people, and the marking content is determined by the types of the abnormal behaviors;
the analysis model module (32) trains according to the labeled tuples of the known personnel to obtain a learning model;
the prediction early warning module (33) predicts the abnormal behavior of the person to be predicted of the person who can not judge whether the person has the abnormal behavior and is forced to be isolated and abstained from the drug taking personnel by using a learning model, and sends alarm and prediction information to the data application unit (4) according to the prediction result of the person to be predicted;
and the data application unit (4) performs linkage response with a security system according to the alarm and prediction information sent by the prediction early warning module (33).
The data information of the person who forcibly isolates the drug addiction comprises the following data information: personal information, health information, family information, consumption information, position information, drug abuse information and abnormal information; the data storage unit (2) stores various data information in a classified manner.
The business system comprises a drug-stopping and law-enforcing comprehensive management platform, an education correction system, a medical health system, a production labor system and a drug-stopping guarantee system; the security system comprises a drug rehabilitation security comprehensive management platform, a video monitoring system, a meeting management system, an access control system, a broadcasting system and an alarm system.
The forced isolation drug rehabilitation personnel comprise: m persons who have been judged to have abnormal behavior and are forced to be isolated for giving up drugs, namely M known persons, and S persons who cannot be judged whether to have abnormal behavior and are forced to be isolated for giving up drugs, namely S persons to be predicted; and the categories of abnormal behaviors of the M known persons are also known, and the categories of the abnormal behaviors comprise: non-polar behavior, escape behavior, suicide behavior, violence behavior, damage behavior, other abnormal behavior;
the feature attribute items include: basic information characteristics, individual evaluation dimension characteristics and drug rehabilitation dynamic event information; wherein the basic information characteristics comprise age, nationality, education degree, previous occupation, drug taking history, strong withdrawal period and drug rehabilitation state; the individual evaluation dimension characteristics comprise camber, clever and sensitive, sympathy, dependency, fluctuation, impulsion, abstinence, self-attitude, anxiety, violence tendency and metamorphosis psychology; the dynamic event information for detoxification includes event type, manual evaluation grade and corresponding score.
The specific way of the numerical preprocessing is as follows:
respectively carrying out numerical preprocessing on the characteristic attributes of M known persons and S persons to be predicted to obtain a tuple X of each known personm，1≤m≤M，Xm＝[Xmj]J is 1,2,3 … d, i.e. Xm＝[Xm1,Xm2,Xm3,…Xmd]And tuple XmEach element X in (1)mjAll correspond to a characteristic attribute item; and obtaining the tuple X of each person to be predicteds，1≤s≤S，Xs＝[Xsj]J is 1,2,3 … d, i.e. Xs＝[Xs1,Xs2,Xs3,…Xsd]And tuple XsEach element in (1)XsjAll correspond to a characteristic attribute item;
wherein m represents the mth known person, XmTuple representing mth known person, j representing jth feature attribute item, XmjThe attribute value of the jth characteristic attribute item representing the mth known person;
s denotes the s-th person to be predicted, XsRepresenting the s-th tuple of the person to be predicted, j representing the j-th characteristic attribute item, XsjThe attribute value of the jth characteristic attribute item representing the s-th person to be predicted;
and the total number of the characteristic attribute items of each known person and each person to be predicted is d, and the number of the characteristic attribute items of each person to be predicted and each known person from the 1 st characteristic attribute item to the end of the d-th characteristic attribute item are kept consistent.
The specific mode of the marking is as follows:
tuple X of known personsmIs added with a mark bit YmObtaining the tuple X 'of the known person after marking'mI.e. X'm＝[Xm1,Xm2,Xm3,…Xmd,Ym]，Ym0, 1,2,3, 4, 5; wherein, YmA category representing abnormal behavior of the mth known person; y ism0 indicates that the category of the abnormal behavior belongs to the electrodeless end behavior; y ism1 indicates that the category of abnormal behavior belongs to escape behavior; y ism2 indicates that the category of abnormal behavior belongs to suicidal behavior; y ismThe category of abnormal behaviors is represented as 3, and belongs to the storm behaviors; y ism4 indicates that the category of abnormal behavior belongs to destructive behavior; y ismThe category of the abnormal behavior is represented by 5, which belongs to other abnormal behaviors.
The analysis model module (32) adopts an ensemble learning algorithm to carry out tuple X 'on the M marked known persons'mTraining is carried out to generate a learning model, namely a behavior characteristic integrated learning model ES, and the specific mode of the integrated learning algorithm is as follows:
m marked known person tuples X'mAs a training set Dtr，Dtr＝{X′m1,2,3 … M (D)tr＝{X′1,X′2,X′3…X′MFor training set DtrM tagged known personnel's tuple X ' of 'mPerforming a replaced random sampling, namely, a hositing sampling to obtain T sample subsets Dtr_tT is 1,2,3, … T, i.e. Dtr_1,Dtr_2,Dtr_3,…,Dtr_T(ii) a And each subset of samples Dtr_tAll contain M tuples X 'of marked known persons'm(ii) a Wherein the subscript tr _ t denotes the t-th, Dtr_tRepresenting the tth sample subset;
adopting a judgment tree algorithm to respectively carry out the judgment on each sample subset Dtr_tPerforming learning training and obtaining the sample subset Dtr_tCorresponding classification model CtT is 1,2,3, … T; i.e. according to the T sample subsets Dtr_tRespectively obtaining T classification models Ct(ii) a Wherein the subscript t denotes the t-th, CtRepresenting the t classification model;
will be classified by T classification models CtThe set of components is a behavior feature integrated learning model ES, which is a learning model { C ═ C1,C2,C3…CT}。
The prediction early warning module (33) is used for predicting the tuples X of S persons to be predictedsAs test set Dts，Ds＝{X1,X2…XSRespectively predicting abnormal behaviors of each person to be predicted in the test set; the specific manner of the abnormal behavior prediction is as follows:
the tuple X of the s-th person to be predictedsRespectively integrating T classification models C in learning model ES through behavior characteristicstCarrying out abnormal behavior prediction and respectively obtaining T predicted values of the s-th person to be predicted t＝1,2,3,…T，s＝1,2,3,…S；
Wherein,the tth predicted value representing the s-th person to be predicted;
the category representing the predicted abnormal behavior belongs to the electrodeless end behavior;a category representing predicted abnormal behavior belongs to escape behavior;a category representing predicted abnormal behavior belongs to suicidal behavior;the category representing the predicted abnormal behavior belongs to the storm behavior;a category representing predicted abnormal behavior belongs to the destructive behavior;the category representing the predicted abnormal behavior belongs to other abnormal behaviors;
counting T predicted valuesIn (1),the number of (a) is v0,the number of (a) is v1,the number of (a) is v2,the number of (a) is v3,the number of (a) is v4,the number of (2) is v 5; comparing the sizes of v0, v1, v2, v3, v4 and v5,
if v0 is maximumIf the number of the people to be predicted is the largest, the prediction result of the s-th person to be predicted is the non-polar behavior without abnormal behavior tendency;
if v1 is maximumIf the number of the person to be predicted is the largest, the prediction result of the s-th person to be predicted is that an abnormal behavior tendency exists, and the type of the predicted abnormal behavior belongs to escape behaviors;
if v2 is maximumIf the number of the abnormal behaviors is the largest, the prediction result of the s-th person to be predicted is that the abnormal behavior tendency exists, and the type of the abnormal behavior is predicted to belong to suicide behavior;
if v3 is maximumIf the number of the predicted persons is the largest, the prediction of the s-th person to be predictedThe test result shows that the abnormal behavior tendency exists, and the category of the abnormal behavior is predicted to belong to the storm behavior;
if v4 is maximumIf the number of the abnormal behaviors is the largest, the prediction result of the s-th person to be predicted is that the abnormal behavior tends to exist, and the type of the predicted abnormal behavior belongs to the destructive behavior;
if v5 is maximumIf the number of the abnormal behaviors is the largest, the prediction result of the s-th person to be predicted is that the abnormal behavior tends to exist, and the type of the predicted abnormal behavior belongs to other abnormal behaviors;
the prediction early warning module (33) gives an alarm to the person to be predicted with the prediction result of abnormal behavior tendency, and sends the prediction information of the person to be predicted with the abnormal behavior tendency, namely the prediction abnormal behavior category, to the data application unit (4).
The data application unit (4) comprises a data query module (41), an alarm linkage module (42), a command scheduling module (43) and a prediction information management module (44): wherein,
the data query module (41) is used for querying various data information in the whole system;
the alarm linkage module (42) realizes linkage response with a related security system according to the alarm and prediction information sent by the prediction early warning module (33);
the command scheduling module (43) realizes unified command scheduling of emergency resources when abnormal behavior events occur;
and the prediction information management module (44) inquires, counts and derives the alarm and prediction information sent by the prediction early warning module (33) for reference of related personnel.
The invention also provides a big data-based abnormal behavior prediction method for the forced isolation of drug addicts, which comprises the following specific steps:
s1, collecting data information of each person from the service system and security system of the mandatory isolation drug rehabilitation department;
s2, storing the data information of each person who forces to isolate drug abstinence;
s3, constructing characteristic attribute items of the forced isolation drug-dropping personnel according to the data information of the forced isolation drug-dropping personnel;
s4, respectively carrying out numerical preprocessing on the characteristic attribute items of the M known persons and the S persons to be predicted to obtain the tuple X of each known personmM is not less than 1 and not more than M, and Xm＝[Xm1,Xm2,Xm3,…Xmd]Tuple XmEach element X in (1)mjAll correspond to a characteristic attribute item; and obtaining the tuple X of each person to be predictedsS is not less than 1 and not more than S, and Xs＝[Xs1,Xs2,Xs3,…Xsd]Tuple XsEach element X in (1)sjAll correspond to a characteristic attribute item;
s5, marking the known person according to the type of the abnormal behavior of the known person, wherein the marking content is the type of the abnormal behavior of the known person; i.e. tuple X at known personmIs added with a mark bit YmObtaining the tuple X 'of the known person after marking'mI.e. X'm＝[Xm1,Xm2,Xm3,…Xmd,Ym]，Ym＝0，1，2，3，4，5；
S6, M labeled known persons tuple X'mAs a training set Dtr，Dtr＝{X′1,X′2,X′3…X′MFor training set DtrM tagged known personnel's tuple X ' of 'mTo carry out with putting backMachine sampling, namely, hositing sampling, to obtain T sample subsets Dtr_tT is 1,2,3, … T, i.e. Dtr_1,Dtr_2,Dtr_3,…,Dtr_T(ii) a And each subset of samples Dtr_tAll contain M tuples X 'of marked known persons'm；
S7, adopting judgment tree algorithm to respectively carry out judgment on each sample subset Dtr_tPerforming learning training and obtaining the sample subset Dtr_tCorresponding classification model CtT is 1,2,3, … T; i.e. according to the T sample subsets Dtr_tRespectively obtaining T classification models Ct(ii) a From T classification models CtThe formed set is a behavior feature integrated learning model ES, wherein ES is { C ═ C1,C2,C3…CT}；
S8, the tuples X of S persons to be predictedsAs test set Dts，Ds＝{X1,X2…XSThe tuple X of each person to be predicted in the test set is respectively treatedsAll the abnormal behaviors are predicted; wherein, the tuple X of the s-th person to be predictedsRespectively and sequentially carrying out abnormal behavior prediction through each classification model in the behavior feature integrated learning model ES, and respectively obtaining T predicted values of the s-th person to be predictedt＝1,2,3,…T，s＝1,2,3,…S；
S9, predicting T predicted valuesThe values of (a) are counted, wherein,the number of (a) is v0,the number of (a) is v1,the number of (a) is v2,the number of (a) is v3,the number of (a) is v4,the number of (2) is v 5; comparing the sizes of v0, v1, v2, v3, v4 and v5, and if v0 is the maximum valueIf the number of the people to be predicted is the largest, the prediction result of the s-th person to be predicted is the non-polar behavior without abnormal behavior tendency; if v1 is maximumIf the number of the person to be predicted is the largest, the prediction result of the s-th person to be predicted is that an abnormal behavior tendency exists, and the type of the predicted abnormal behavior belongs to escape behaviors; if v2 is maximumIf the number of the abnormal behaviors is the largest, the prediction result of the s-th person to be predicted is that the abnormal behavior tendency exists, and the type of the abnormal behavior is predicted to belong to suicide behavior; if v3 is maximumIf the number of the abnormal behaviors is the largest, the prediction result of the s-th person to be predicted is that the abnormal behavior tends to exist, and the type of the abnormal behavior is predicted to belong to the storm behavior; if v4 is maximumIf the number of the abnormal behaviors is the largest, the prediction result of the s-th person to be predicted is that the abnormal behavior tends to exist, and the type of the predicted abnormal behavior belongs to the destructive behavior; if v5 is maximumIf the number of the abnormal behaviors is the largest, the prediction result of the s-th person to be predicted is that the abnormal behavior tends to exist, and the type of the predicted abnormal behavior belongs to other abnormal behaviors;
and S10, alarming the person to be predicted with the prediction result of abnormal behavior tendency.
The invention has the advantages that:
(1) the invention adopts big data technology, acquires data from the business system and the security system of the forced isolation drug rehabilitation institute, constructs a plurality of characteristic attribute items of the forced isolation drug rehabilitation personnel, makes prediction more comprehensive, and provides guarantee for the accuracy of the prediction result.
(2) According to the data of the forced isolation drug-dropping personnel judged to have abnormal behaviors, the behavior characteristic integrated learning model of the forced isolation drug-dropping personnel is established, so that the abnormal behaviors of the forced isolation drug-dropping personnel are accurately predicted, and the defects of the traditional mode are overcome.
(3) The behavior feature integrated learning model comprises a plurality of classification models, the abnormal behaviors of the person who forcibly isolates the drug addicts are respectively predicted through the classification models, and the obtained classification results are fused, so that the defect of poor system stability of only one classification model is overcome, and the classification precision and stability of the system are effectively improved.
(4) The invention subdivides the types of the predicted abnormal behaviors, and predicts the types of the abnormal behaviors through the behavior characteristic integrated learning model, thereby being beneficial to policemen to adopt targeted measures to prevent the abnormal behaviors from happening, and improving the safety prevention level of the forced isolation drug rehabilitation institute.
(5) The invention classifies and stores the data information of the person who forcibly isolates the drug rehabilitation, records the prediction information, facilitates the inquiry and statistics of the supervision personnel and improves the supervision level of the person who forcibly isolates the drug rehabilitation.
(6) The invention has stronger self-adaptive capacity, high prediction accuracy on the abnormal behavior of the forced isolation drug-dropping personnel and small influence caused by misjudgment.
Drawings
Fig. 1 is an overall architecture diagram of the big data-based abnormal behavior prediction system for forced isolation of drug addicts according to the present invention.
Fig. 2 is a flow chart of a prediction method of the big data-based abnormal behavior prediction system for the person who breaks the drug addiction by force isolation.
Fig. 3 is a schematic diagram of a prediction method of the abnormal behavior prediction system for the person who is forced to be isolated from drug addiction based on big data.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, the system for predicting abnormal behavior of a person who is forced to be isolated from drug addiction based on big data comprises the following components: the system comprises a data acquisition unit 1, a data storage unit 2, a data analysis unit 3 and a data application unit 4;
the data acquisition unit 1 is connected with the business system and the security system of the forced isolation drug rehabilitation institute, and acquires data information of the forced isolation drug rehabilitation personnel from the business system and the security system of the forced isolation drug rehabilitation institute. Wherein,
the business system comprises a drug-stopping and law-enforcing comprehensive management platform, an education correction system, a medical health system, a production labor system and a drug-stopping guarantee system;
the security system comprises a drug rehabilitation security comprehensive management platform, a video monitoring system, a meeting management system, an access control system, a broadcasting system and an alarm system;
the data information of the person who forcibly isolates the drug addiction comprises the following data information: personal information, health information, family information, consumption information, location information, drug abuse information, and exception information.
The data storage unit 2 is used for storing various data information of the person who forcibly isolates the drug addiction treatment in a classified manner.
The data analysis unit 3 includes a data preprocessing module 31, an analysis model module 32, and a prediction and early warning module 33. Wherein,
the data preprocessing module 31 constructs a characteristic attribute item of the person who forcibly isolates the drug addiction according to various data information of the person who forcibly isolates the drug addiction, wherein the characteristic attribute item comprises: basic information characteristics, individual evaluation dimension characteristics and drug rehabilitation dynamic event information; the basic information characteristics comprise age, nationality, education degree, previous occupation, drug taking history, strong abstinence period and drug abstinence state; the individual evaluation dimension characteristics comprise camber, clever and sensitive, sympathy, dependency, fluctuation, impulsion, abstinence, self-attitude, anxiety, violence tendency and metamorphosis psychology; the drug-dropping dynamic event information comprises an event type, a manual evaluation grade and a corresponding score;
the forced isolation drug rehabilitation personnel comprise: m persons who have been judged to have abnormal behavior and are forced to be isolated for giving up drugs, namely M known persons, and S persons who cannot be judged whether to have abnormal behavior and are forced to be isolated for giving up drugs, namely S persons to be predicted; and the categories of abnormal behaviors of the M known persons are also known, and the categories of the abnormal behaviors comprise: non-polar behavior, escape behavior, suicide behavior, violence behavior, damage behavior, other abnormal behavior;
the data preprocessing module 31 respectively performs numerical preprocessing on the feature attribute items of the M known persons and the S persons to be predicted to obtain a tuple X of each known personm，1≤m≤M，Xm＝[Xmj]J is 1,2,3 … d, i.e. Xm＝[Xm1,Xm2,Xm3,…Xmd]And tuple XmEach element X in (1)mjAll correspond to a characteristic attribute item; and obtaining the tuple X of each person to be predicteds，1≤s≤S，Xs＝[Xsj]J is 1,2,3 … d, i.e. Xs＝[Xs1,Xs2,Xs3,…Xsd]And tuple XsEach element X in (1)sjAll correspond to a characteristic attribute item;
wherein m represents the mth known person, XmTuple representing mth known person, j representing jth feature attribute item, XmjThe attribute value of the jth characteristic attribute item representing the mth known person;
s denotes the s-th person to be predicted, XsRepresenting the s-th tuple of the person to be predicted, j representing the j-th characteristic attribute item, XsjThe attribute value of the jth characteristic attribute item representing the s-th person to be predicted;
the total number of the characteristic attribute items of each known person and each person to be predicted is d, and the number of the characteristic attribute items of each person to be predicted and each known person from the 1 st characteristic attribute item to the end of the d-th characteristic attribute item are kept consistent;
the data preprocessing module 31 further marks the known person according to the category of the abnormal behavior of the known person, and the marking content is the category of the abnormal behavior of the known person; the specific mode of the mark is as follows: tuple X of known personsmIs added with a mark bit YmObtaining the tuple X 'of the known person after marking'mI.e. X'm＝[Xm1,Xm2,Xm3,…Xmd,Ym]，Ym＝0，1，2，3，45, 5; wherein, YmA category representing abnormal behavior of the mth known person; y ism0 indicates that the category of the abnormal behavior belongs to the electrodeless end behavior; y ism1 indicates that the category of abnormal behavior belongs to escape behavior; y ism2 indicates that the category of abnormal behavior belongs to suicidal behavior; y ismThe category of abnormal behaviors is represented as 3, and belongs to the storm behaviors; y ism4 indicates that the category of abnormal behavior belongs to destructive behavior; y ismThe category of the abnormal behavior is represented by 5, which belongs to other abnormal behaviors.
The analysis model module 32 employs a ensemble learning algorithm to pair the tuples X 'of the M labeled known persons'mTraining is carried out, a behavior characteristic ensemble learning model ES is generated, and the concrete mode of the ensemble learning algorithm is as follows:
m marked known person tuples X'mAs a training set Dtr，Dtr＝{X′m1,2,3 … M (D)tr＝{X′1,X′2,X′3…X′MFor training set DtrM tagged known personnel's tuple X ' of 'mPerforming a replaced random sampling, namely, a hositing sampling to obtain T sample subsets Dtr_tT is 1,2,3, … T, i.e. Dtr_1,Dtr_2,Dtr_3,…,Dtr_T(ii) a And each subset of samples Dtr_tAll contain M tuples X 'of marked known persons'm(ii) a Wherein the subscript tr _ t denotes the t-th, Dtr_tRepresenting the tth sample subset;
adopting a judgment tree algorithm to respectively carry out the judgment on each sample subset Dtr_tPerforming learning training and obtaining the sample subset Dtr_tCorresponding classification model CtT is 1,2,3, … T; i.e. according to the T sample subsets Dtr_tRespectively obtaining T classification models Ct(ii) a Wherein the subscript t denotes the t-th, CtRepresenting the t classification model; the judgment tree algorithm can be referred to in the prior art specifically;
from T classification models CtComposition ofThe set of (a) is a behavior feature integrated learning model ES, wherein ES is { C }1,C2,C3…CT}。
The prediction and early warning module 33 uses the behavior feature integrated learning model ES to respectively predict the abnormal behavior of each person to be predicted, and sends alarm and prediction information to the data application unit 4 according to the prediction result of the person to be predicted, specifically as follows:
the tuples X of S persons to be predictedsAs test set Dts，Ds＝{X1,X2…XSRespectively carrying out abnormal behavior prediction on the tuple of each person to be predicted in the test set; the abnormal behavior is predicted by the tuple X of the s-th person to be predictedsRespectively and sequentially carrying out abnormal behavior prediction through each classification model in the behavior feature integrated learning model ES, and respectively obtaining T predicted values of the s-th person to be predictedT is 1,2,3, … T, S is 1,2,3, … S; wherein,the tth predicted value representing the s-th person to be predicted;
and isWherein,the category representing the predicted abnormal behavior belongs to the electrodeless end behavior;a category representing predicted abnormal behavior belongs to escape behavior;the category representing the predicted abnormal behavior belongs toKilling the behavior;the category representing the predicted abnormal behavior belongs to the storm behavior;a category representing predicted abnormal behavior belongs to the destructive behavior;the category representing the predicted abnormal behavior belongs to other abnormal behaviors;
counting T predicted valuesIn (1),the number of (a) is v0,the number of (a) is v1,the number of (a) is v2,the number of (a) is v3,the number of (a) is v4,the number of (2) is v 5; comparing the sizes of v0, v1, v2, v3, v4 and v5, and if v0 is the maximum valueIf the number of the people to be predicted is the largest, the prediction result of the s-th person to be predicted is the non-polar behavior without abnormal behavior tendency; if v1 is maximumIf the number of the person to be predicted is the largest, the prediction result of the s-th person to be predicted is that an abnormal behavior tendency exists, and the type of the predicted abnormal behavior belongs to escape behaviors; if v2 is maximumIf the number of the abnormal behaviors is the largest, the prediction result of the s-th person to be predicted is that the abnormal behavior tendency exists, and the type of the abnormal behavior is predicted to belong to suicide behavior; if v3 is maximumIf the number of the abnormal behaviors is the largest, the prediction result of the s-th person to be predicted is that the abnormal behavior tends to exist, and the type of the abnormal behavior is predicted to belong to the storm behavior; if v4 is maximumIf the number of the abnormal behaviors is the largest, the prediction result of the s-th person to be predicted is that the abnormal behavior tends to exist, and the type of the predicted abnormal behavior belongs to the destructive behavior; if v5 is maximumIf the number of the abnormal behaviors is the largest, the prediction result of the s-th person to be predicted is that the abnormal behavior tends to exist, and the type of the predicted abnormal behavior belongs to other abnormal behaviors;
the prediction early warning module 33 gives an alarm to the person to be predicted whose prediction result is that the person to be predicted has the abnormal behavior tendency, and sends the prediction information, i.e., the prediction abnormal behavior category, of the person to be predicted having the abnormal behavior tendency to the data application unit 4.
When the prediction early warning module 33 sends an alarm, the data application unit 4 realizes linkage response with a related security system, realizes unified command and scheduling of emergency resources, and can inquire, count and export prediction information for reference of related personnel; the data application unit 4 includes a data query module 41, an alarm linkage module 42, a command scheduling module 43, and a prediction information management module 44. Wherein,
the data query module 41 is configured to query various data information in the entire system;
the alarm linkage module 42 realizes linkage response with related security systems according to the alarm and prediction information sent by the prediction early warning module 33;
when an abnormal behavior event occurs, the command scheduling module 43 realizes unified command scheduling of emergency resources;
the prediction information management module 44 queries, counts and derives the alarm and prediction information sent by the prediction and early warning module 33 for reference by the relevant personnel.
As shown in fig. 2 and 3, the prediction method of the abnormal behavior prediction system for the forced isolation drug addict based on big data comprises the following steps:
s1, the data acquisition unit 1 acquires the data information of each person in the forced isolation drug rehabilitation from the business system and the security system of the forced isolation drug rehabilitation institute;
s2, the data storage unit 2 stores each item of data information of each person who forces to isolate drug abstinence;
s3, the data preprocessing module 31 constructs the characteristic attribute item of the person who forces to isolate the drug addiction according to the data information of the person who forces to isolate the drug addiction;
s4, the data preprocessing module 31 respectively carries out numerical preprocessing on the characteristic attribute items of M forced isolation drug addicts (M known people) judged to have abnormal behaviors and S forced isolation drug addicts (S people to be predicted) which can not judge whether to have abnormal behaviors, and the tuple X of each known person is obtainedmM is not less than 1 and not more than M, and Xm＝[Xmj]J is 1,2,3 … d, i.e. Xm＝[Xm1,Xm2,Xm3,…Xmd]Tuple XmEach element X in (1)mjAll correspond to a characteristic attribute item; and obtaining the tuple X of each person to be predictedsS is not less than 1 and not more than S, and Xs＝[Xsj]J is 1,2,3 … d, i.e. Xs＝[Xs1,Xs2,Xs3,…Xsd]Tuple XsEach element X in (1)sjAll correspond to a characteristic attribute item;
s5, the data preprocessing module 31 marks the known person according to the type of the abnormal behavior of the known person, and the marking content is the type of the abnormal behavior of the known person; i.e. tuple X at known personmIs added with a mark bit YmObtaining the tuple X 'of the known person after marking'mI.e. X'm＝[Xm1,Xm2,Xm3,…Xmd,Ym]，Ym＝0，1，2，3，4，5；
S6, the analysis model module 32 maps the M labeled tuples of known people X'mAs a training set Dtr，Dtr＝{X′m1,2,3 … M (D)tr＝{X′1,X′2,X′3…X′MFor training set DtrM tagged known personnel's tuple X ' of 'mPerforming a replaced random sampling, namely, a hositing sampling to obtain T sample subsets Dtr_tT is 1,2,3, … T, i.e. Dtr_1,Dtr_2,Dtr_3,…,Dtr_T(ii) a And each subset of samples Dtr_tAll contain M tuples X 'of marked known persons'm；
S7, the analysis model module 32 adopts the decision tree algorithm to respectively determine each sample subset Dtr_tPerforming learning training and obtaining the sample subset Dtr_tCorresponding classification model CtT is 1,2,3, … T; i.e. according to the T sample subsets Dtr_tRespectively obtaining T classification models Ct(ii) a From T classification models CtThe formed set is a behavior feature integrated learning model ES, wherein ES is { C ═ C1,C2,C3…CT}；
S8, the prediction early-warning module 33 will predict the tuples X of S people to be predictedsAs test set Dts，Ds＝{X1,X2…XSThe tuple X of each person to be predicted in the test set is respectively treatedsAll the abnormal behaviors are predicted; the abnormal behavior prediction is as follows: the tuple X of the s-th person to be predictedsRespectively and sequentially carrying out abnormal behavior prediction through each classification model in the behavior feature integrated learning model ES, and respectively obtaining T predicted values of the s-th person to be predictedt＝1,2,3,…T，s＝1,2,3,…S；
S9, the prediction early warning module 33 also predicts T predicted valuesThe values of (a) are counted, wherein,the number of (a) is v0,the number of (a) is v1,the number of (a) is v2,the number of (a) is v3,the number of (a) is v4,the number of (2) is v 5; comparing the sizes of v0, v1, v2, v3, v4 and v5, and if v0 is the maximum valueIf the number of the people to be predicted is the largest, the prediction result of the s-th person to be predicted is the non-polar behavior without abnormal behavior tendency; if v1 is maximumIf the number of the person to be predicted is the largest, the prediction result of the s-th person to be predicted is that an abnormal behavior tendency exists, and the type of the predicted abnormal behavior belongs to escape behaviors; if v2 is maximumIf the number of the abnormal behaviors is the largest, the prediction result of the s-th person to be predicted is that the abnormal behavior tendency exists, and the type of the abnormal behavior is predicted to belong to suicide behavior; if v3 is maximumIf the number of the abnormal behaviors is the largest, the prediction result of the s-th person to be predicted is that the abnormal behavior tends to exist, and the type of the abnormal behavior is predicted to belong to the storm behavior; if v4 is maximumIf the number of the abnormal behaviors is the largest, the prediction result of the s-th person to be predicted is that the abnormal behavior tends to exist, and the type of the predicted abnormal behavior belongs to the destructive behavior; if v5 is maximumIf the number of the abnormal behaviors is the largest, the prediction result of the s-th person to be predicted is that the abnormal behavior tends to exist, and the type of the predicted abnormal behavior belongs to other abnormal behaviors;
s10, the prediction and early warning module 33 alarms the person to be predicted whose prediction result is the person with abnormal behavior tendency, and sends the prediction information of the person to be predicted with abnormal behavior tendency, that is, the prediction abnormal behavior category to the data application unit 4.
The behavior feature integrated learning model of the forced isolation drug-dropping personnel is constructed by adopting a big data technology, so that the abnormal behavior of the forced isolation drug-dropping personnel is accurately predicted, and the defects of the traditional modes such as manual experience evaluation and the like are overcome; the behavior characteristic integrated learning model of the forced isolation drug-dropping person based on big data overcomes the defect of poor system stability of one classification model by generating a plurality of differential classification models and fusing the classification results of the differential classification models, and effectively improves the classification precision and stability of the system. Therefore, the invention has stronger self-adaptive capacity, higher prediction accuracy on the abnormal behavior of the person who forcibly isolates the drug abstinence, and smaller influence caused by misjudgment.
The invention is not to be considered as limited to the specific embodiments shown and described, but is to be understood to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the invention as defined by the appended claims.
Claims (10)
1. The abnormal behavior prediction system of the person who forces to keep apart drug addiction based on big data, characterized by that, the said system includes: the device comprises a data acquisition unit (1), a data storage unit (2), a data analysis unit (3) and a data application unit (4);
the data acquisition unit (1) is connected with a business system and a security system of the forced isolation drug rehabilitation institute, and acquires data information of the forced isolation drug rehabilitation personnel from the business system and the security system of the forced isolation drug rehabilitation institute; the data acquisition unit (1) stores the acquired data information of the forced isolation drug rehabilitation personnel to the data storage unit (2);
the data analysis (3) unit comprises a data preprocessing module (31), an analysis model module (32) and a prediction early warning module (33); wherein,
the data preprocessing module (31) acquires data information of the person who forcibly isolates the drug to be stopped from the data storage unit (2), and constructs characteristic attribute items of the person who forcibly isolates the drug to be stopped according to the data information of the person who forcibly isolates the drug to be stopped from the drug to be stopped; the data preprocessing module (31) carries out numerical preprocessing on the characteristic attribute items of the forced isolation drug-dropping personnel to obtain tuples of the forced isolation drug-dropping personnel; the data preprocessing module (31) also marks tuples of the forced isolation drug addicts, namely known people, which are judged to have abnormal behaviors, so as to obtain the marked tuples of the known people, and the marking content is determined by the types of the abnormal behaviors;
the analysis model module (32) trains according to the labeled tuples of the known personnel to obtain a learning model;
the prediction early warning module (33) predicts the abnormal behavior of the person to be predicted of the person who can not judge whether the person has the abnormal behavior and is forced to be isolated and abstained from the drug taking personnel by using a learning model, and sends alarm and prediction information to the data application unit (4) according to the prediction result of the person to be predicted;
and the data application unit (4) performs linkage response with a security system according to the alarm and prediction information sent by the prediction early warning module (33).
2. The big data based abnormal behavior prediction system of a person who forces to quarantine drugs according to claim 1, wherein the data information of the person who forces to quarantine drugs includes: personal information, health information, family information, consumption information, position information, drug abuse information and abnormal information; the data storage unit (2) stores various data information in a classified manner.
3. The system for predicting abnormal behavior of a person who is forced to be isolated from drug addiction based on big data according to claim 1, wherein the business system comprises a drug addiction enforcement comprehensive management platform, an education correction system, a medical health system, a production labor system and a drug addiction guarantee system; the security system comprises a drug rehabilitation security comprehensive management platform, a video monitoring system, a meeting management system, an access control system, a broadcasting system and an alarm system.
4. The big data based abnormal behavior prediction system of a person with forced drug abstinence according to claim 1, wherein the person with forced drug abstinence comprises: m persons who have been judged to have abnormal behavior and are forced to be isolated for giving up drugs, namely M known persons, and S persons who cannot be judged whether to have abnormal behavior and are forced to be isolated for giving up drugs, namely S persons to be predicted; and the categories of abnormal behaviors of the M known persons are also known, and the categories of the abnormal behaviors comprise: non-polar behavior, escape behavior, suicide behavior, violence behavior, damage behavior, other abnormal behavior;
the feature attribute items include: basic information characteristics, individual evaluation dimension characteristics and drug rehabilitation dynamic event information; wherein the basic information characteristics comprise age, nationality, education degree, previous occupation, drug taking history, strong withdrawal period and drug rehabilitation state; the individual evaluation dimension characteristics comprise camber, clever and sensitive, sympathy, dependency, fluctuation, impulsion, abstinence, self-attitude, anxiety, violence tendency and metamorphosis psychology; the dynamic event information for detoxification includes event type, manual evaluation grade and corresponding score.
5. The big-data-based system for predicting abnormal behavior of a person abstaining from drugs according to claim 4, wherein the numerical preprocessing is performed in the following manner:
respectively carrying out numerical preprocessing on the characteristic attributes of M known persons and S persons to be predicted to obtain a tuple X of each known personm，1≤m≤M，Xm＝[Xmj]J is 1,2,3 … d, i.e. Xm＝[Xm1,Xm2,Xm3,…Xmd]And tuple XmEach element X in (1)mjAll correspond to a characteristic attribute item; and obtaining the tuple X of each person to be predicteds，1≤s≤S，Xs＝[Xsj]J is 1,2,3 … d, i.e. Xs＝[Xs1,Xs2,Xs3,…Xsd]And tuple XsEach element X in (1)sjAll correspond to a characteristic attribute item;
wherein m represents the mth known person, XmTuple representing mth known person, j representing jth feature attribute item, XmjThe attribute value of the jth characteristic attribute item representing the mth known person;
s denotes the s-th person to be predicted, XsRepresenting the s-th tuple of the person to be predicted, j representing the j-th characteristic attribute item, XsjThe attribute value of the jth characteristic attribute item representing the s-th person to be predicted;
and the total number of the characteristic attribute items of each known person and each person to be predicted is d, and the number of the characteristic attribute items of each person to be predicted and each known person from the 1 st characteristic attribute item to the end of the d-th characteristic attribute item are kept consistent.
6. The big data based abnormal behavior prediction system of a person who is forced to be abstained from drugs according to claim 5, wherein the specific manner of marking is as follows:
tuple X of known personsmIs added with a mark bit YmObtaining the tuple X 'of the known person after marking'mI.e. X'm＝[Xm1,Xm2,Xm3,…Xmd,Ym]，Ym0, 1,2,3, 4, 5; wherein, YmA category representing abnormal behavior of the mth known person; y ism0 indicates that the category of the abnormal behavior belongs to the electrodeless end behavior; y ism1 indicates that the category of abnormal behavior belongs to escape behavior; y ism2 indicates that the category of abnormal behavior belongs to suicidal behavior; y ismThe category of abnormal behaviors is represented as 3, and belongs to the storm behaviors; y ism4 indicates that the category of abnormal behavior belongs to destructive behavior; y ismThe category of the abnormal behavior is represented by 5, which belongs to other abnormal behaviors.
7. The big-data based abnormal behavior prediction system for forced quarantine abstaining personnel of claim 6, wherein the analysis model module (32) employs an ensemble learning algorithm to tuple X 'of M labeled known personnel'mTraining is carried out to generate a learning model, namely a behavior characteristic integrated learning model ES, and the specific mode of the integrated learning algorithm is as follows:
m marked known person tuples X'mAs a training set Dtr，Dtr＝{X′m1,2,3 … M (D)tr＝{X′1,X′2,X′3…X′MFor training set DtrM tagged known personnel's tuple X ' of 'mPerforming a replaced random sampling, namely, a hositing sampling to obtain T sample subsets Dtr_tT is 1,2,3, … T, i.e. Dtr_1,Dtr_2,Dtr_3,…,Dtr_T(ii) a And each subset of samples Dtr_tAll contain M tuples X 'of marked known persons'm(ii) a Wherein the subscript tr _ t denotes the t-th, Dtr_tRepresenting the tth sample subset;
adopting a judgment tree algorithm to respectively carry out the judgment on each sample subset Dtr_tPerforming learning training and obtaining the sample subset Dtr_tCorresponding classification model CtT is 1,2,3, … T; i.e. according to the T sample subsets Dtr_tRespectively obtaining T classification models Ct(ii) a Wherein the subscript t denotes the t-th, CtRepresenting the t classification model;
will be classified by T classification models CtThe set of components is a behavior feature integrated learning model ES, which is a learning model { C ═ C1,C2,C3…CT}。
8. The big data based abnormal behavior prediction system of a person abstaining from drugs according to claim 7, wherein the system is characterized byThe prediction early warning module (33) is used for predicting the tuples X of S persons to be predictedsAs test set Dts，Ds＝{X1,X2…XSRespectively predicting abnormal behaviors of each person to be predicted in the test set; the specific manner of the abnormal behavior prediction is as follows:
the tuple X of the s-th person to be predictedsRespectively integrating T classification models C in learning model ES through behavior characteristicstCarrying out abnormal behavior prediction and respectively obtaining T predicted values of the s-th person to be predicted
Wherein,the tth predicted value representing the s-th person to be predicted;
the category representing the predicted abnormal behavior belongs to the electrodeless end behavior;a category representing predicted abnormal behavior belongs to escape behavior;a category representing predicted abnormal behavior belongs to suicidal behavior;the category representing the predicted abnormal behavior belongs to the storm behavior;indication deviceDetecting that the category of the abnormal behavior belongs to a destructive behavior;the category representing the predicted abnormal behavior belongs to other abnormal behaviors;
counting T predicted valuesIn (1),the number of (a) is v0,the number of (a) is v1,the number of (a) is v2,the number of (a) is v3,the number of (a) is v4,the number of (2) is v 5; comparing the sizes of v0, v1, v2, v3, v4 and v5,
if v0 is maximumIf the number of the people to be predicted is the largest, the prediction result of the s-th person to be predicted is the non-polar behavior without abnormal behavior tendency;
if v1 is maximumIf the number of the predicted persons is the largest, the prediction result of the s-th person to be predicted is presentThe abnormal behaviors tend, and the category of the abnormal behaviors is predicted to belong to escape behaviors;
if v2 is maximumIf the number of the abnormal behaviors is the largest, the prediction result of the s-th person to be predicted is that the abnormal behavior tendency exists, and the type of the abnormal behavior is predicted to belong to suicide behavior;
if v3 is maximumIf the number of the abnormal behaviors is the largest, the prediction result of the s-th person to be predicted is that the abnormal behavior tends to exist, and the type of the abnormal behavior is predicted to belong to the storm behavior;
if v4 is maximumIf the number of the abnormal behaviors is the largest, the prediction result of the s-th person to be predicted is that the abnormal behavior tends to exist, and the type of the predicted abnormal behavior belongs to the destructive behavior;
if v5 is maximumIf the number of the abnormal behaviors is the largest, the prediction result of the s-th person to be predicted is that the abnormal behavior tends to exist, and the type of the predicted abnormal behavior belongs to other abnormal behaviors;
the prediction early warning module (33) gives an alarm to the person to be predicted with the prediction result of abnormal behavior tendency, and sends the prediction information of the person to be predicted with the abnormal behavior tendency, namely the prediction abnormal behavior category, to the data application unit (4).
9. The big data-based abnormal behavior prediction system for forced isolation of drug addicts according to claim 2, wherein the data application unit (4) comprises a data query module (41), an alarm linkage module (42), a commanding and scheduling module (43), and a prediction information management module (44): wherein,
the data query module (41) is used for querying various data information in the whole system;
the alarm linkage module (42) realizes linkage response with a related security system according to the alarm and prediction information sent by the prediction early warning module (33);
the command scheduling module (43) realizes unified command scheduling of emergency resources when abnormal behavior events occur;
and the prediction information management module (44) inquires, counts and derives the alarm and prediction information sent by the prediction early warning module (33) for reference of related personnel.
10. The abnormal behavior prediction method of the person who abstains from drugs in forced isolation based on big data is characterized by comprising the following specific steps:
s1, collecting data information of each person from the service system and security system of the mandatory isolation drug rehabilitation department;
s2, storing the data information of each person who forces to isolate drug abstinence;
s3, constructing characteristic attribute items of the forced isolation drug-dropping personnel according to the data information of the forced isolation drug-dropping personnel;
s4, respectively carrying out numerical preprocessing on the characteristic attribute items of the M known persons and the S persons to be predicted to obtain the tuple X of each known personmM is not less than 1 and not more than M, and Xm＝[Xm1,Xm2,Xm3,…Xmd]Tuple XmEach element X in (1)mjAll correspond to a characteristic attribute item; and obtaining the tuple X of each person to be predictedsS is not less than 1 and not more than S, and Xs＝[Xs1,Xs2,Xs3,…Xsd]Tuple XsEach element X in (1)sjAll correspond to a characteristic attribute item;
s5, marking the known person according to the type of the abnormal behavior of the known person, wherein the marking content is the type of the abnormal behavior of the known person; i.e. tuple X at known personmIn addition to oneMarker bit YmObtaining the tuple X 'of the known person after marking'mI.e. X'm＝[Xm1,Xm2,Xm3,…Xmd,Ym]，Ym＝0，1，2，3，4，5；
S6, M labeled known persons tuple X'mAs a training set Dtr，Dtr＝{X′1,X′2,X′3…X′MFor training set DtrM tagged known personnel's tuple X ' of 'mPerforming a replaced random sampling, namely, a hositing sampling to obtain T sample subsets Dtr_tT is 1,2,3, … T, i.e. Dtr_1,Dtr_2,Dtr_3,…,Dtr_T(ii) a And each subset of samples Dtr_tAll contain M tuples X 'of marked known persons'm；
S7, adopting judgment tree algorithm to respectively carry out judgment on each sample subset Dtr_tPerforming learning training and obtaining the sample subset Dtr_tCorresponding classification model CtT is 1,2,3, … T; i.e. according to the T sample subsets Dtr_tRespectively obtaining T classification models Ct(ii) a From T classification models CtThe formed set is a behavior feature integrated learning model ES, wherein ES is { C ═ C1,C2,C3…CT}；
S8, the tuples X of S persons to be predictedsAs test set Dts，Ds＝{X1,X2…XSThe tuple X of each person to be predicted in the test set is respectively treatedsAll the abnormal behaviors are predicted; wherein, the tuple X of the s-th person to be predictedsRespectively and sequentially carrying out abnormal behavior prediction through each classification model in the behavior feature integrated learning model ES, and respectively obtaining T predicted values of the s-th person to be predicted
S9, predicting T predicted valuesThe values of (a) are counted, wherein,the number of (a) is v0,the number of (a) is v1,the number of (a) is v2,the number of (a) is v3,the number of (a) is v4,the number of (2) is v 5; comparing the sizes of v0, v1, v2, v3, v4 and v5, and if v0 is the maximum valueIf the number of the people to be predicted is the largest, the prediction result of the s-th person to be predicted is the non-polar behavior without abnormal behavior tendency; if v1 is maximumIf the number of the person to be predicted is the largest, the prediction result of the s-th person to be predicted is that an abnormal behavior tendency exists, and the type of the predicted abnormal behavior belongs to escape behaviors; if v2 is maximumWhen the number of (2) is the maximum, thenThe prediction results of the s persons to be predicted show that abnormal behaviors tend to exist, and the types of the abnormal behaviors are predicted to belong to suicide behaviors; if v3 is maximumIf the number of the abnormal behaviors is the largest, the prediction result of the s-th person to be predicted is that the abnormal behavior tends to exist, and the type of the abnormal behavior is predicted to belong to the storm behavior; if v4 is maximumIf the number of the abnormal behaviors is the largest, the prediction result of the s-th person to be predicted is that the abnormal behavior tends to exist, and the type of the predicted abnormal behavior belongs to the destructive behavior; if v5 is maximumIf the number of the abnormal behaviors is the largest, the prediction result of the s-th person to be predicted is that the abnormal behavior tends to exist, and the type of the predicted abnormal behavior belongs to other abnormal behaviors;
and S10, alarming the person to be predicted with the prediction result of abnormal behavior tendency.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811451746.9A CN109523088A (en) | 2018-11-30 | 2018-11-30 | The abnormal behaviour forecasting system of forced quarantine addict received treatment based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811451746.9A CN109523088A (en) | 2018-11-30 | 2018-11-30 | The abnormal behaviour forecasting system of forced quarantine addict received treatment based on big data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109523088A true CN109523088A (en) | 2019-03-26 |
Family
ID=65794562
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811451746.9A Pending CN109523088A (en) | 2018-11-30 | 2018-11-30 | The abnormal behaviour forecasting system of forced quarantine addict received treatment based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109523088A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110020687A (en) * | 2019-04-10 | 2019-07-16 | 北京神州泰岳软件股份有限公司 | Abnormal behaviour analysis method and device based on operator's Situation Awareness portrait |
| CN111159210A (en) * | 2019-12-31 | 2020-05-15 | 北京明略软件系统有限公司 | Information processing system, method and device |
| CN112598184A (en) * | 2020-12-27 | 2021-04-02 | 上海达梦数据库有限公司 | Method and device for predicting relapse risk of drug-dropping person |
| CN113077197A (en) * | 2021-06-08 | 2021-07-06 | 泰豪信息技术有限公司 | Escort personnel consumption supervision method, system, storage medium and equipment |
-
2018
- 2018-11-30 CN CN201811451746.9A patent/CN109523088A/en active Pending
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110020687A (en) * | 2019-04-10 | 2019-07-16 | 北京神州泰岳软件股份有限公司 | Abnormal behaviour analysis method and device based on operator's Situation Awareness portrait |
| CN110020687B (en) * | 2019-04-10 | 2021-11-05 | 北京神州泰岳软件股份有限公司 | Abnormal behavior analysis method and device based on operator situation perception portrait |
| CN111159210A (en) * | 2019-12-31 | 2020-05-15 | 北京明略软件系统有限公司 | Information processing system, method and device |
| CN111159210B (en) * | 2019-12-31 | 2023-06-13 | 北京明智和术科技有限公司 | Information processing system, method and device |
| CN112598184A (en) * | 2020-12-27 | 2021-04-02 | 上海达梦数据库有限公司 | Method and device for predicting relapse risk of drug-dropping person |
| CN112598184B (en) * | 2020-12-27 | 2024-02-02 | 上海达梦数据库有限公司 | Method and device for predicting repeated air suction risk of drug addict |
| CN113077197A (en) * | 2021-06-08 | 2021-07-06 | 泰豪信息技术有限公司 | Escort personnel consumption supervision method, system, storage medium and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109523088A (en) | The abnormal behaviour forecasting system of forced quarantine addict received treatment based on big data | |
| CN104182805B (en) | Dangerous tendency Forecasting Methodology based on inmate's behavioural characteristic integrated study model | |
| Markson et al. | Linking serial residential burglary: Comparing the utility of modus operandi behaviours, geographical proximity, and temporal proximity | |
| Dempster et al. | The relative utility of fixed and variable risk factors in discriminating sexual recidivists and nonrecidivists | |
| Goebel | Social unrest in China: a bird’s-eye view | |
| CN108881194A (en) | Enterprises user anomaly detection method and device | |
| CN106295565A (en) | Monitor event identifications based on big data and in real time method of crime prediction | |
| Porter | A statistical approach to crime linkage | |
| Pyrooz et al. | Gang organization, offending, and victimization: A cross-national analysis | |
| Tonkin et al. | Behavioural case linkage with solved and unsolved crimes | |
| Meade | The labeling approach to delinquency: State of the theory as a function of method | |
| Sherman et al. | The felonious few vs. the miscreant many | |
| CN110826852B (en) | Risk assessment method and system for forced isolation drug-relief personnel | |
| Cole et al. | Behavioural investigative advice: Assistance to investigative decision‐making in difficult‐to‐detect murder | |
| van Schendel | The challenges of risk profiling used by law enforcement: Examining the cases of COMPAS and SyRI | |
| CN114003683B (en) | Alarm analysis method based on natural language processing and association rules | |
| Willmott et al. | Criminal geography and geographical profiling within police investigations–a brief introduction | |
| CN107196942A (en) | A kind of inside threat detection method based on user language feature | |
| Lundrigan et al. | Factors predicting conviction in child stranger rape | |
| Yokota et al. | Crime linkage of sex offences in Japan by multiple correspondence analysis | |
| Weatherburn et al. | Screening juvenile offenders for further assessment and intervention | |
| Paine | Solvability factors in dwelling burglaries in Thames Valley | |
| CN116910662A (en) | Passenger anomaly identification method and device based on random forest algorithm | |
| Fridel et al. | The quantitative study of serial murder: Regression is not transgression | |
| Peterson et al. | How often and under what circumstances do escapes from prison and jail result in violence? An examination of the “powder keg” theory |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190326 |
|
| RJ01 | Rejection of invention patent application after publication |