CN114997897A - Mobile data-based method for constructing images of easily-damaged people - Google Patents
Mobile data-based method for constructing images of easily-damaged people Download PDFInfo
- Publication number
- CN114997897A CN114997897A CN202210362270.1A CN202210362270A CN114997897A CN 114997897 A CN114997897 A CN 114997897A CN 202210362270 A CN202210362270 A CN 202210362270A CN 114997897 A CN114997897 A CN 114997897A
- Authority
- CN
- China
- Prior art keywords
- user
- space
- data
- feature
- information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 27
- 230000004927 fusion Effects 0.000 claims abstract description 13
- 238000012216 screening Methods 0.000 claims abstract description 5
- 238000004364 calculation method Methods 0.000 claims description 21
- 238000013499 data model Methods 0.000 claims description 6
- 238000012217 deletion Methods 0.000 claims description 3
- 230000037430 deletion Effects 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 2
- 239000000126 substance Substances 0.000 claims description 2
- 238000001914 filtration Methods 0.000 claims 1
- 230000005251 gamma ray Effects 0.000 claims 1
- 238000007418 data mining Methods 0.000 abstract description 3
- 238000005516 engineering process Methods 0.000 abstract description 3
- 230000006399 behavior Effects 0.000 description 14
- 238000011160 research Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
-
- 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
- G06Q30/00—Commerce
- G06Q30/018—Certifying business or products
- G06Q30/0185—Product, service or business identity fraud
-
- 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0203—Market surveys; Market polls
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Game Theory and Decision Science (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a method for constructing a vulnerable group portrait based on mobile data, and belongs to the field of data mining. The method comprises the following steps: s1: extracting spatial features of user network information; s2: extracting user physical space behavior characteristics; s3: acquiring user binary space fusion characteristics, namely performing characteristic fusion on user network information space characteristics and physical space behavior characteristics, and screening the user binary space fusion characteristics to obtain an optimal characteristic subset; s4: a user representation is generated from the optimal feature subset. The method combines the data capacity of a mobile operator, the data of a third-party electronic map, the user portrait technology and the characteristics of victims in telecommunication fraud, constructs the user portrait of the easily-injured people, and effectively improves the accuracy and integrity of the portrait. The data is utilized to deepen the understanding of the victim, and a new idea is provided for the effective development of anti-fraud work.
Description
Technical Field
The invention belongs to the field of data mining, and relates to a method for constructing an image of a vulnerable crowd based on mobile data.
Background
Position data, internet surfing data and social data left by the mobile phone in the using process can reflect living habits and social modes of people to a certain extent. At present, in the field of telecommunication anti-fraud, a user portrait is often constructed aiming at data of a fraudulent party, and researches on a victim are generally carried out by only utilizing basic attributes of the victim, such as age, sex and the like, or researches on the psychology of the victim in a manner of writing and the like for specific cases. The current research neglects the analysis of the binary space behavior characteristics of the victim, so the invention provides the image concept of the easily-attacked people and the construction method of the easily-attacked people, which not only can improve the management efficiency of the information behaviors of the easily-attacked people, but also is beneficial for the related personnel to understand the easily-attacked people to formulate an accurate anti-fraud strategy.
Disclosure of Invention
In view of the above, the present invention provides a method for constructing a vulnerable group portrait based on mobile data, which combines mobile communication big data with a user portrait technology to analyze behavior characteristics of a victim, improve accuracy and integrity of portrait, facilitate formulation of accurate anti-fraud measures, and optimize a range of propaganda objects, thereby realizing anti-fraud.
In order to achieve the purpose, the invention provides the following technical scheme:
a method for constructing a vulnerable group portrait based on mobile data specifically comprises the following steps:
s1: extracting the spatial characteristics of user network information;
s2: extracting user physical space behavior characteristics;
s3: acquiring user binary space fusion characteristics, namely performing characteristic fusion on user network information space characteristics and physical space behavior characteristics, and screening the user binary space fusion characteristics to obtain an optimal characteristic subset;
s4: a user representation is generated from the optimal feature subset.
Further, in step S1, the extracting spatial features of user network information specifically includes the following steps:
s11: extracting basic attribute information, call record information, short message record information and internet access record information of the user by using a related data model in a database associating the mobile phone number of the telecom fraud user with the mobile phone number of the normal user;
s12: and using the digital mark user basic information to summarize the conversation behavior, the short message behavior and the internet surfing behavior according to the time dimension.
Further, in step S2, extracting the user physical space behavior feature specifically includes the following steps:
s21: associating the mobile phone number of the telecommunication fraud user with the mobile phone number of the normal user with a relevant data model in a database, extracting data of a user access base station and longitude and latitude information of the base station, and introducing POI information by using a third-party electronic map;
s22: identifying a user's stay point according to the interval weight of two adjacent points in the user trajectory data and the local space-time density of the trajectory data;
s23: each stopping point is endowed with semantic information, and the most POI types in the radius R of the stopping point are endowed to the stopping point, wherein the POI types comprise 12 types of residential areas, companies, catering areas, shopping areas, living areas, scenic areas, lodging areas, automobile services, financial areas, cultural areas, sports services and medical services;
s24: the number of times the user visits each POI-type area is summarized by time.
Further, in step S22, the interval weight γ c The calculation formula of (2) is as follows:
wherein, the first and the second end of the pipe are connected with each other,the distance interval normalized value of two adjacent position points is represented,representing the time interval normalization value of two adjacent position points;
the local space-time density ρ i The calculation formula of (2) is as follows:
wherein, dist (x) i ,x j ) Is x i To x j Ground distance of (x) i 、x j Respectively representing two position points in the trajectory, t i 、t j Respectively representing the time stamps of two position points in a track, wherein n represents the total number of data points in one track, and the sigma value is set as 1;
the method for identifying the stop point comprises the following steps: for gamma c Performing a Kmeans algorithm with a clustering cluster number k of 2 into two classes, and dividing gamma c Gamma of larger value class c Minimum value as gamma c_th Setting the interval weight value less than gamma c_th Dividing all points between two adjacent deletion positions into a candidate stop point, and then pairing rho in each candidate stop point i The Kmeans algorithm, which performs k 2, is grouped into two classes, p i Rho of a less valued class i Maximum value as ρ i_th Presence of more than ρ in the candidate dwell points i_th The space-time density of (a) is then the actual dwell point.
Further, in step S3, feature fusion is performed on the user network information space feature and the physical space behavior feature, and the calculation formula is:
wherein, a, b, c are different types of characteristics in user network information space, x is a certain type of characteristics in user physical space, subscript letter is corresponding type characteristic quantity, and x is Cartesian product.
Further, in step S3, the method for screening the user binary space fusion features specifically includes the following steps:
s31: calculating the maximum mutual information coefficient MIC of each feature and the target class in the original feature space; the calculation formula of the maximum mutual information coefficient MIC is as follows:
wherein D { (f) 1 ,C),(f 2 ,C),...,(f i C) is a feature f i And log (min (X, Y)) is a normalized value of mutual information quantity, XY represents that a data space is divided by using a grid of X × Y, and XY is less than 0.6 th power of total data quantity; in the data space after the grid division, the calculation formula of the frequency P (x, y) of the data points falling in the (x, y) -th grid is as follows:
I * (D, X, Y) represents the size of the mutual information amount in the case where X and Y are specified, and the calculation formula is:
deleting the features smaller than the threshold epsilon, and then sorting in a descending order to form a feature subset S;
s32: calculating the symmetry uncertainty SU between the features and the target categories; the calculation formula of the symmetry uncertainty SU is as follows:
wherein, I (f) i (ii) a C) Representation feature f i And mutual information of class C, H (f) i ) And H (C) respectively represent the feature f i The information entropy of (1) and the information entropy of class C; characteristic f i Is a characteristic f j The approximate markov blanket's conditional expression is:
SU(f i ,C)>SU(f j ,C)&SU(f i ,f j )>SU(f j ,C)
wherein SU (f) i C) represents the feature f i And classCorrelations between other C, SU (f) i ,f j ) Representing a feature f i And characteristic f j The correlation between them; and finally, deleting redundant features to generate an optimal feature subset F.
Further, in step S4, generating a user profile according to the optimal feature subset specifically includes the following steps:
s41: representing all features in the optimal feature subset as fact labels;
s42: constructing a plurality of feature labels according to the fact labels and the psychology of the victim in the telecommunication fraud, including but not limited to information exposure degree, credit facility degree, social contact degree and financial attributes; the calculation formula of the characteristic label is as follows:
wherein f is j Representing optimal feature subset and feature tag V i Normalized value of the associated feature, n being the value associated with the distinctive mark V i The number of features that are relevant.
The invention has the beneficial effects that:
1) the invention combines the big data capacity of mobile operators, third-party data resources, user portrait technology and the characteristics of victims in telecom fraud, constructs the user portrait in a complete binary space of a user by utilizing a data mining algorithm, improves the accuracy and the integrity of the portrait, and provides a new idea for anti-fraud work.
2) The invention introduces a two-stage feature selection algorithm to respectively remove irrelevant and redundant features, so that an optimal feature subset is reserved, and the features most relevant to the vulnerable group of people can be effectively extracted.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention may be realized and attained by the means of the instrumentalities and combinations particularly pointed out hereinafter.
Drawings
For a better understanding of the objects, aspects and advantages of the present invention, reference will now be made to the following detailed description taken in conjunction with the accompanying drawings in which:
FIG. 1 is a general flow chart of the method of the present invention for building a picture of a vulnerable group of people based on mobile data;
FIG. 2 is a flow chart of a feature selection portion of the method of the present invention;
fig. 3 is an example of the correlation between the victim attribute tags and the underlying features of the present invention.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention in a schematic way, and the features in the following embodiments and examples may be combined with each other without conflict.
Referring to fig. 1 to fig. 3, the present invention provides a method for constructing a drawing of a vulnerable group of people based on mobile data, as shown in the general flowchart of fig. 1, the method specifically includes the following steps:
s1: selecting the materials in a ratio of 1: 1, extracting basic attribute information, call record information, short message record information and internet access record information of the user by using a relevant data model in a database associating the mobile phone number of the telecommunication fraud user with the mobile phone number of the normal user.
S2: and using the digital mark user basic information to summarize the conversation behavior, the short message behavior and the internet surfing behavior according to the time dimension.
S3: and associating the mobile phone number of the victim user with the mobile phone number of the normal user with a relevant data model in a database, extracting the data of the user access base station and the latitude and longitude information of the base station, and introducing POI information by using a third-party electronic map.
S4: identifying a user's stay point according to the interval weight of two adjacent points in the user trajectory data and the local space-time density of the trajectory data; the calculation formula of the interval weight in the stay point detection algorithm is as follows:
wherein the content of the first and second substances,the distance interval normalized value of two adjacent position points is represented,and the time interval normalized value of two adjacent position points is represented.
The local space-time density calculation formula is expressed as:
wherein, dist (x) i ,x j ) Is x i To x j Ground distance of (x) i 、x j Respectively representing two position points in the trajectory, t i 、t j Respectively representing time stamps of two position points in a track, wherein n represents the total number of data points in one track, and the sigma value is set as 1;
the method for identifying the stop point comprises the following steps: for gamma c Performing a Kmeans algorithm with a clustering cluster number k of 2 into two classes, and dividing gamma c Gamma of larger value class c Minimum value as gamma c_th Make the interval weight less than gamma c_th Dividing all points between two adjacent deletion positions into a candidate stop point, and then pairing rho in each candidate stop point i The Kmeans algorithm, which performs k 2, is grouped into two classes, p i Rho of a less valued class i Maximum value as ρ i_th Presence of more than ρ in the candidate stopping point i_th The space-time density of (a) is then the actual dwell point.
S5: and each stopping point is endowed with semantic information, and the most POI types in the radius R of the stopping point are endowed to the stopping point, wherein the POI types comprise 12 types of residential areas, companies, catering areas, shopping areas, living areas, scenic areas, lodging areas, automobile services, financial areas, cultural areas, sports services and medical services.
S6: the number of times the user visits each POI-type area is summarized by time.
S7: carrying out feature fusion on the user network information space features and the physical space behavior features; the feature crossing process under the user binary space is as follows:
wherein, a, b, c are different types of characteristics in user network information space, x is a certain type of characteristics in user physical space, subscript letter is corresponding type characteristic quantity, and x is Cartesian product.
As shown in fig. 2, the process of feature screening for features in the user binary space includes:
s8: calculating the maximum mutual information coefficient MIC of each feature and target class in the original feature space, wherein the maximum mutual information coefficient MIC is expressed as:
wherein D { (f) 1 ,C),(f2,C),...,(f i C) is a feature f i And log (min (X, Y)) is a normalized value of mutual information quantity, XY represents that a data space is divided by using a grid of X × Y, and XY is less than 0.6 th power of total data quantity; in the gridded data space, the frequency P (x, y) of data points falling in the (x, y) -th grid is calculated as:
I * (D, X, Y) represents the mutual information amount size in the case of specifying X and Y, and the calculation formula is:
deleting the features smaller than the threshold epsilon, and then sorting in a descending order to form a feature subset S;
s9: calculating the symmetry uncertainty SU between the features and the characteristics and between the features and the target category, wherein the calculation formula of the symmetry uncertainty SU is as follows:
wherein, I (f) i (ii) a C) Representation feature f i And mutual information of class C, H (f) i ) And H (C) respectively represent the feature f i And the information entropy of class C. Characteristic f i Is a characteristic f j The approximate markov blanket's conditional expression is:
SU(f i ,C)>SU(f j ,C)&SU(f i ,f j )>SU(f j ,C)
wherein SU (f) i C) represents the feature f i Correlation with class C, SU (f) i ,f j ) Representing a feature f i And characteristic f j The correlation between them; and finally, deleting redundant features to generate an optimal feature subset F.
The process of generating a user representation includes:
s10: all features in the optimal feature subset are represented as fact labels;
s11: several feature tags are constructed according to the fact tags in combination with the psychology of the victim in the telecommunication fraud, including but not limited to information exposure degree, credit worthiness degree, social contact degree, financial attributes. The calculation formula of the characteristic label is as follows:
wherein, f j Representing optimal feature subset and feature tag V i The normalized value of the related characteristic is calculated by the formula:
wherein, X norm Is normalized data, X is raw data, X max 、X min Respectively, the maximum and minimum values in the raw data. n is a special label V i The number of features that are relevant. As FIG. 3 shows the basic features associated with the feature tag, Table 1 shows the related insights of the basic features and the feature tag.
TABLE 1 relevant insights of basic features and feature labels
Finally, the above embodiments are only intended to illustrate the technical solutions of the present invention and not to limit the present invention, and although the present invention has been described in detail with reference to the preferred embodiments, it will be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions, and all of them should be covered by the claims of the present invention.
Claims (7)
1. A method for constructing an image of a vulnerable group of people based on mobile data is characterized by comprising the following steps:
s1: extracting the spatial characteristics of user network information;
s2: extracting user physical space behavior characteristics;
s3: acquiring user binary space fusion characteristics, namely performing characteristic fusion on user network information space characteristics and physical space behavior characteristics, and screening the user binary space fusion characteristics to obtain an optimal characteristic subset;
s4: a user representation is generated from the optimal feature subset.
2. The method of claim 1, wherein the step of extracting spatial characteristics of user network information in step S1 comprises the following steps:
s11: extracting basic attribute information, call record information, short message record information and internet access record information of the user by using a related data model in a database associating the mobile phone number of the telecom fraud user with the mobile phone number of the normal user;
s12: and using the digital mark user basic information to summarize the conversation behavior, the short message behavior and the internet surfing behavior according to the time dimension.
3. The method as claimed in claim 1, wherein the step of extracting the user physical space behavior characteristics in step S2 specifically comprises the steps of:
s21: associating the mobile phone number of the telecom fraud user with the mobile phone number of the normal user with a relevant data model in a database, extracting data of a user access base station and longitude and latitude information of the base station, and introducing POI information by using a third-party electronic map;
s22: identifying a dwell point of a user according to an interval weight of two adjacent points in the user track data and the local space-time density of the track data;
s23: assigning semantic information to each stop point, and assigning the most POI types in the radius R of the stop point to the stop point;
s24: the number of times the user visits each POI-type area is summarized by time.
4. The method as claimed in claim 3, wherein the interval weight γ 22 c The calculation formula of (2) is as follows:
wherein the content of the first and second substances,the distance interval normalized value of two adjacent position points is represented,representing the time interval normalization values of two adjacent position points;
the local space-time density ρ i The calculation formula of (2) is as follows:
wherein, dist (x) i ,x j ) Is x i To x j Ground distance of (x) i 、x j Respectively representing two position points in the trajectory, t i 、t j Respectively representing time stamps of two position points in a track, wherein n represents the total number of data points in one track, and the sigma value is set as 1;
the method for identifying the stop point comprises the following steps: for gamma ray c Performing a Kmeans algorithm with a clustering cluster number k of 2 into two classes, and dividing gamma c Gamma of larger value class c Minimum value as gamma c_th Setting the interval weight value less than gamma c_th Dividing all points between two adjacent deletion positions into a candidate stop point, and then pairing rho in each candidate stop point i The Kmeans algorithm, which performs k 2, is grouped into two classes, p i Rho of a less valued class i Maximum value as ρ i_th Presence of more than ρ in the candidate dwell points i_th The space-time density of (a) is then the actual dwell point.
5. The method of claim 1, wherein in step S3, the user network information space characteristic and the physical space behavior characteristic are feature fused, and the calculation formula is:
wherein, a, b, c are different types of characteristics in the user network information space, x is a certain type of characteristics in the user physical space, subscript letter is the corresponding type characteristic quantity, and x is Cartesian product.
6. The method for constructing a portrait of people subject to fraud as claimed in claim 1, wherein the step S3 of filtering the user binary space fusion features comprises the following steps:
s31: calculating the maximum mutual information coefficient MIC of each feature and the target class in the original feature space; the calculation formula of the maximum mutual information coefficient MIC is as follows:
wherein D { (f) 1 ,C),(f 2 ,C),...,(f i C) is a feature f i And log (min (X, Y)) is a normalized value of mutual information quantity, XY represents that a data space is divided by using a grid of X × Y, and XY is less than 0.6 th power of total data quantity; in the data space after the grid division, the calculation formula of the frequency P (x, y) of the data points falling in the (x, y) -th grid is as follows:
I * (D, X, Y) represents the mutual information amount size in the case of specifying X and Y, and the calculation formula is:
deleting the features smaller than the threshold epsilon, and then sorting in a descending order to form a feature subset S;
s32: calculating the symmetry uncertainty SU between the features and the target category; the calculation formula of the symmetry uncertainty SU is as follows:
wherein, I (f) i (ii) a C) Representing a feature f i And mutual information of class C, H (f) i ) And H (C) respectively represent the feature f i Information entropy of (d) and information entropy of class C; characteristic f i Is a characteristic f j The approximate markov blanket's conditional expression is:
SU(f i ,C)>SU(f j ,C)&SU(f i ,f j )>SU(f j ,C)
wherein SU (f) i C) represents the feature f i And correlation between class C, SU (f) i ,f j ) Representation feature f i And characteristic f j The correlation between them; and finally, deleting redundant features to generate an optimal feature subset F.
7. The method of claim 1, wherein the step of generating a user representation according to the optimal feature subset in step S4 comprises the following steps:
s41: representing all features in the optimal feature subset as fact labels;
s42: constructing a plurality of characteristic labels according to the fact labels and the psychology of the victim in the telecommunication fraud; the calculation formula of the characteristic label is as follows:
wherein f is j Representing optimal feature subset and feature tag V i Normalized value of the associated feature, n being associated with the distinctive tag V i Number of related features。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210362270.1A CN114997897A (en) | 2022-04-07 | 2022-04-07 | Mobile data-based method for constructing images of easily-damaged people |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210362270.1A CN114997897A (en) | 2022-04-07 | 2022-04-07 | Mobile data-based method for constructing images of easily-damaged people |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114997897A true CN114997897A (en) | 2022-09-02 |
Family
ID=83024013
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210362270.1A Pending CN114997897A (en) | 2022-04-07 | 2022-04-07 | Mobile data-based method for constructing images of easily-damaged people |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114997897A (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP3461291A1 (en) * | 2017-09-09 | 2019-04-03 | Apple Inc. | Implementation of biometric authentication |
CN111144452A (en) * | 2019-12-11 | 2020-05-12 | 重庆邮电大学 | Mobile user trip chain extraction method based on signaling data and clustering algorithm |
CN111340121A (en) * | 2020-02-28 | 2020-06-26 | 支付宝(杭州)信息技术有限公司 | Target feature determination method and device |
CN111753520A (en) * | 2020-06-02 | 2020-10-09 | 五八有限公司 | Risk prediction method and device, electronic equipment and storage medium |
CN112925989A (en) * | 2021-01-29 | 2021-06-08 | 中国计量大学 | Group discovery method and system of attribute network |
-
2022
- 2022-04-07 CN CN202210362270.1A patent/CN114997897A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP3461291A1 (en) * | 2017-09-09 | 2019-04-03 | Apple Inc. | Implementation of biometric authentication |
CN111144452A (en) * | 2019-12-11 | 2020-05-12 | 重庆邮电大学 | Mobile user trip chain extraction method based on signaling data and clustering algorithm |
CN111340121A (en) * | 2020-02-28 | 2020-06-26 | 支付宝(杭州)信息技术有限公司 | Target feature determination method and device |
CN111753520A (en) * | 2020-06-02 | 2020-10-09 | 五八有限公司 | Risk prediction method and device, electronic equipment and storage medium |
CN112925989A (en) * | 2021-01-29 | 2021-06-08 | 中国计量大学 | Group discovery method and system of attribute network |
Non-Patent Citations (2)
Title |
---|
顾金哲: "基于通信大数据的易受诈人群画像研究", 《中国优秀硕士学位论文全文数据库 社会科学Ⅰ辑》, 15 June 2023 (2023-06-15), pages 113 - 12 * |
黄先奇: "基于改进随机森林的异常移动通讯用户的检测研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》, 15 March 2022 (2022-03-15), pages 136 - 156 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108681936B (en) | Fraud group identification method based on modularity and balanced label propagation | |
WO2020238631A1 (en) | Population type recognition method based on mobile phone signaling data | |
CN107527240B (en) | System and method for identifying public praise marketing effect of operator industry product | |
Kim | Exploring the difference between ridership patterns of subway and taxi: Case study in Seoul | |
CN109635056B (en) | Power utilization address data processing method and device, computer equipment and storage medium | |
CN104615616A (en) | Group recommendation method and system | |
US20210263957A1 (en) | Method and apparatus for dividing region, storage medium, and electronic device | |
CN111080501B (en) | Real crowd density space-time distribution estimation method based on mobile phone signaling data | |
CN107016042B (en) | Address information verification system based on user position log | |
CN106951828B (en) | Urban area function attribute identification method based on satellite images and network | |
CN113613174A (en) | Method, device and storage medium for identifying occupational sites based on mobile phone signaling data | |
CN112668375B (en) | Tourist distribution analysis system and method in scenic spot | |
CN113010578B (en) | Community data analysis method and device, community intelligent interaction platform and storage medium | |
Manley et al. | New forms of data for understanding urban activity in developing countries | |
CN112258254A (en) | Internet advertisement risk monitoring method and system based on big data architecture | |
CN108898244B (en) | Digital signage position recommendation method coupled with multi-source elements | |
Rodrigues et al. | Automatic classification of points-of-interest for land-use analysis | |
CN110008402A (en) | A kind of point of interest recommended method of the decentralization matrix decomposition based on social networks | |
CN110288468B (en) | Data feature mining method and device, electronic equipment and storage medium | |
CN112258250A (en) | Target user identification method and device based on network hotspot and computer equipment | |
CN114693317A (en) | Telecommunication fraud security federation detection method fusing homogeneous graph and bipartite graph | |
Wang et al. | Influence of built environment on subway trip origin and destination: insights based on mobile positioning data | |
CN111639251A (en) | Information retrieval method and device | |
CN114997897A (en) | Mobile data-based method for constructing images of easily-damaged people | |
CN116017333A (en) | Population identification method, system and storage medium based on big data signaling processing |
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 |