CN110245981B - Crowd type identification method based on mobile phone signaling data - Google Patents
Crowd type identification method based on mobile phone signaling data Download PDFInfo
- Publication number
- CN110245981B CN110245981B CN201910467120.5A CN201910467120A CN110245981B CN 110245981 B CN110245981 B CN 110245981B CN 201910467120 A CN201910467120 A CN 201910467120A CN 110245981 B CN110245981 B CN 110245981B
- Authority
- CN
- China
- Prior art keywords
- mobile phone
- crowd
- signaling data
- phone signaling
- feature
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- 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
Abstract
The invention discloses a crowd type identification method based on mobile phone signaling data, and belongs to the technical field of crowd type identification. The invention combines the mobile phone signaling data with the basic attribute information of the mobile phone user to mine and extract the relevant characteristics of the crowd trip. And (4) sorting the features according to the importance degree by utilizing a backward elimination method through calculating the total distance entropy among all samples so as to select the features. Based on the screened features, a k-means clustering method is utilized to perform clustering analysis on the mobile phone signaling data, and clustering clusters are divided. And identifying the crowd type of each cluster by combining the distribution condition of each crowd in the corresponding characteristics. Compared with the prior art, the method can more fully mine the information in the mobile phone signaling data, and analyze the category attributes of the crowd from the global perspective by utilizing a machine learning method. The method not only reduces the dependence and the requirement on the prior experience knowledge, improves the applicability of the method, but also can avoid the subjectivity brought by a rule discrimination method.
Description
Technical Field
The invention belongs to the technical field of crowd type identification, and particularly relates to a crowd type identification method based on mobile phone signaling data.
Background
The development and the popularization of communication and internet technologies have brought forward the research and the mining of large data of a large amount of potential information hidden behind the communication and internet technologies. According to the statistics of the ministry of industry and belief, the number of mobile phone users in the whole country reaches 12.86 hundred million by 2015, and the popularization rate of the mobile phone is 94.5 per hundred. The mobile phone signaling data attracts the attention of a large number of scholars due to the excellent characteristics of rich connotation, high sampling rate, good timeliness and the like. However, due to privacy protection and some limitations of the data acquisition field, it is difficult to obtain accurate sample information with a tag, such as type attribute and travel mode of a traveler, which will hinder further promotion of related research on traffic analysis, traffic planning, and the like.
The concept of user portrayal was originally proposed by the parent Cooper of interactive design as an interactive design tool to facilitate and consolidate user-centric design ideas. User portrayal is an important tool in implementing user-centric interactive design as an important component of user research. Through the user portrait, a design team can pay attention to the user and the requirement of the user at any time in the product and service design process, so that the design team can be in consensus with the user. In the traffic field, a traveler is a user in a traffic system, and people type identification is carried out by carrying out user portrait research on the traveler, so that technical support can be provided for people topic research in traffic planning. The process of building a user representation is essentially a process of describing virtual users in short text, i.e., abstracting user features into phrase tags, where the virtual users within each group have similar goals, needs, behaviors, and the like. The short text involved in this description is called portrait description. There are two types of user portrait construction processes in the existing research: one is that the product designer and the operator abstract typical users from user groups according to user requirements; the other type is to generate a label set describing the user according to the data of the behavior, the view and the like of each user in the product, the service and the like. The former portrait is essentially a tool describing the needs of the user, and is used for helping different designers to stand at the user's perspective to think about the problem in the product and service design process. The latter portrait is essentially a tagged user model, which is used to portray user intent. It is clear that the former is heavily and lightly analyzed, and the latter is heavily and lightly analyzed.
The clustering method has a long research history in various scientific fields, k-means, and is proposed for the first time in 1955 as one of the most popular and most common clustering algorithms. In sixty years after the k-means algorithm is proposed, thousands of clustering algorithms are proposed, but the k-means algorithm is still widely used. The algorithm is a mature and effective label-free sample classification method, and samples are grouped together by measuring the similarity among the sample essences. In the aspect of research on mobile phone signaling data, relevant researchers apply the research to travel pattern recognition of travelers, and no researcher applies the research to crowd type recognition of travelers at present.
In the crowd type identification based on the mobile phone signaling data, most of the crowd type identification is realized through a rule discrimination method. The identification method needs to determine the activity rule of the crowd and needs to determine the value range of the corresponding characteristics when distinguishing different types of crowds. The selection of the thresholds needs abundant prior knowledge and has strong subjectivity. Meanwhile, the method has poor expansibility, is sensitive to special dates and research areas, and has poor transportability.
At present, Machun et al disclose a floating population identification method based on mobile phone signaling data. The method comprises the following steps: 1) each user is taken as a research unit, mobile phone signaling data of the user in one day are extracted and are arranged according to a time sequence; 2) dividing a research area into a central urban area, a research urban area and a research provincial area, and assigning an attribute value field attribute to each area; 3) defining floating population, and then further refining and classifying the floating population according to the movement rule of the mobile phone signaling data between the regions; 4) a rule algorithm for judging the floating population is formulated according to the movement rule among the regions; 5) the identification and statistics of different types of floating population are realized by using Java programming. According to the scheme, only mobile phone signaling data of a user in one day is extracted, floating population is identified from the dimension of spatial distribution, and the influence of the time dimension on population identification is ignored.
Disclosure of Invention
The existing crowd type identification method based on mobile phone signaling data generally only uses a rule method of simple logic judgment to identify and extract a certain target crowd during analysis, the set rules have strong subjectivity and need abundant prior experience knowledge, and the accuracy of model identification depends on the establishment of the rules to a great extent. Meanwhile, the identification method for the single target population is poor in extensibility, and due to the fact that the difference of the characteristics required to be selected for dividing different types of populations is large, the identification of the target population under other research purposes is difficult to reproduce. Aiming at the existing defects, the invention provides a crowd type identification method based on mobile phone signaling data.
Based on the mobile phone signaling data, the invention extracts the relevant characteristics of the people going out and the basic attribute information of the user from the overall view of the data sample so as to enhance the applicability of the method to different types of people. And then screening out an optimal feature set by a distance entropy method to remove redundant features, dividing the crowd by using the selected features and applying a k-means clustering analysis method of unsupervised machine learning, and finally carrying out crowd type identification on the clustering cluster by combining feature distribution of each crowd. The crowd is divided and identified by using the machine learning method, the information contained in the mobile phone signaling data is utilized to the maximum extent, the requirement on prior experience knowledge is reduced, the human intervention is reduced, and the objectivity of the method is improved.
The technical scheme of the invention is as follows: a crowd type identification method based on mobile phone signaling data comprises the following steps:
and S1, acquiring the mobile phone signaling data and the corresponding mobile phone user basic attribute information in the research time period of the research area.
And S2, extracting the relevant characteristics of the crowd trip based on the mobile phone signaling data.
And S3, forming a feature set by the crowd travel related features extracted in the step S2 and the basic attribute information of the mobile phone users, and screening out an optimal feature subset by using a distance entropy method.
And S4, carrying out crowd division by using an unsupervised k-means cluster analysis method according to the characteristic subset selected in the step S3.
And S5, carrying out crowd type identification on each cluster according to the characteristic distribution condition of each crowd.
Further, in step S1, the cell phone signaling data includes a cell phone identification code, a timestamp, an event type, a base station number, a base station longitude and latitude, and a number attribution; the basic attribute information of the mobile phone user comprises age, gender, number attribution and household registration location.
Further, step S2 includes the following steps:
s21, traversing the data set, and respectively counting the number of days of the mobile phone user in the research time range;
s22, determining working time intervals and residence time intervals according to prior experience knowledge and by combining actual conditions of research areas, and identifying places of employment, namely residence and working places, according to the residence time of each residence in the intervals of the mobile phone user; if there is no lingering plot that satisfies the condition, the occupational plot is noted as 0. The working time interval refers to a time range of normal working time of a mobile phone user, for example, a working time interval obtained from nine am to five pm according to prior experience knowledge.
And S23, respectively counting the days of the mobile phone users in the residence places, the days of the mobile phone users in the workplace, the days of the mobile phone users in the residence places and the days of the mobile phone users in the workplace based on the identified places of employment.
Further, step S3 includes the following steps:
s31, traversing the data set, and calculating Euclidean distance between each sample, sample xiAnd xjHas an Euclidean distance D betweenijThe calculation formula is as follows (where maxkAnd minkRespectively representing the maximum and minimum values of the kth feature, and M representing the number of features). Note that: traversing a data set is simply a fingerExtracting each data in the data set, wherein the sample refers to mobile phone signaling data of 13 days in Kun mountain city, Jiangsu province, the third table is a table head of the data set, and each row of data below the table head is a sample.
S32, calculating similarity measurement between samples, normalizing the similarity measurement to be between 0 and 1, and continuously changing the sample into variable xiAnd xjSimilarity between SijThe calculation formula is as follows (wherein alpha is a control parameter, the attenuation property of the control similarity is theoretically taken
Wherein
As average distance between objects):
the similarity calculation formula among the discrete variables is as follows, wherein M is the number of discrete variables:
s33, calculating the distance entropy between each sample, sample xiAnd xjEntropy E of the distance betweenijThe calculation formula is as follows:
Eij＝-Sij logSij-(1-Sij)*log(1-Sij)
s34, calculating the distance entropy of the sample population, wherein the calculation formula is as follows, wherein N represents the number of samples:
s35, traversing the feature set by adopting a backward elimination method as a search strategy, calculating the total distance entropy of removing one feature each time, wherein the feature removed corresponding to the total distance entropy with the minimum value is the least important feature, and then removing the feature from the feature set and putting the feature into a new feature set; and repeating the process for one iteration until all the features are transferred into the new feature set, and reversely ordering the new feature set to obtain the importance ranking of the features.
Further, the flow of step S4 is
S41, randomly selecting k samples from the sample set as an initial mean vector [ mu ]1,μ2,...,μk}；
S42, traversing the sample set, calculating the Euclidean distance between each sample x and each mean vector, and calculating the Euclidean distance according to the mean vector mu closest to the sample xiDetermining its cluster mark Ci；
S43, after all samples are divided, recalculating the mean vector, wherein the calculation formula is as follows:
and S44, repeating the step S42 until the current mean vector is not updated.
S45, output cluster division C ═ C1,C2,...,Ck}。
Further, step S5 includes
S51, carrying out crowd detailed classification according to the crowd activity characteristic difference and the research precision requirement;
s52, determining approximate spatial distribution of each characteristic of different types of people;
and S53, judging the crowd type of each cluster by combining the crowd characteristic distribution.
Has the advantages that:
the invention provides a crowd type identification method based on mobile phone signaling data, which is characterized by extracting and selecting crowd trip related characteristics according to the mobile phone signaling data, then dividing crowds by using a k-means unsupervised clustering analysis method, and identifying the crowd type by combining the distribution of the crowds on the related characteristics. Compared with the prior art, the method can more fully mine the information in the mobile phone signaling data, and analyze the category attributes of the crowd from the global perspective by utilizing a machine learning method. The method not only reduces the dependence and the requirement on the prior experience knowledge, improves the applicability of the method, but also can avoid the subjectivity brought by a rule discrimination method.
Drawings
FIG. 1 is a schematic flow diagram of the present invention;
FIG. 2 is a thermodynamic diagram of population distribution in an example;
fig. 3 is an exemplary diagram of three types of people group division based on mobile phone signaling data in the embodiment.
Detailed Description
The present invention is further illustrated by the following figures and specific examples, which are to be understood as illustrative only and not as limiting the scope of the invention, which is to be given the full breadth of the appended claims and any and all equivalent modifications thereof which may occur to those skilled in the art upon reading the present specification.
Example 1
In step S1, the data of the mobile phone signaling of 13 days in kunshan city, jiangsu province is taken as an example, and the mobile phone users appearing in kunshan city during this period are 5863054 users. The basic attribute information of the mobile phone user includes age, gender, number attribution, household registration and a mobile phone unique identifier, as shown in table 1.
TABLE 1 basic Attribute information for Mobile phone subscribers
In step S2, based on the mobile phone signaling data, relevant features of the people going out are extracted. Types and definitions are shown in table 2, and a converged statistical representation based on population travel characteristics is shown in table 3:
TABLE 2 types and definitions of relevant characteristics of people going out
TABLE 3 example of aggregated statistics based on crowd travel characteristics
In step S3, the features are sorted according to their importance by using a distance entropy method. In this embodiment, taking a feature set composed of EXISTS _ DAYS, ULD, UWD, ON _ LSD and ON _ WSD as an example, the calculated features are ON _ WSD, UWD, EXISTS _ DAYS, ON _ LSD and ULD in the order of high to low importance, and the calculation is shown in table 4:
table 4 example of feature selection based on distance entropy method
In step S4, the network graph is divided by using a k-means cluster analysis algorithm based ON EXISTS _ DAYS, UWD and ON _ WSD, the crowd distribution thermodynamic diagram based ON EXISTS _ DAYS and UWD is shown in fig. 2, and the crowd division result is shown in fig. 3 (the number of cluster centers is set to 3, wherein the number of blocks 4242272, the number of circles 600820 and the number of triangles 1019962 are set).
In step S5, the population is roughly divided into a transit population, a floating population and a resident population, and considering that the resident population lives in the study area in most of the time within the study time range, the frequency of leaving the study area is relatively low, and EXISTS _ DAYS, UWD and ON _ WSD are large; the floating population is generally a middle-long term business trip population or a cross-border trip population, the trip behavior is regular in a period of time, and EXISTS _ DAYS, UWD and ON _ WSD are relatively large; the transit population generally stays in the research area for a short time, leaves the area frequently, and EXISTS _ DAYS, UWD and ON _ WSD are small. From this, it was concluded that the square is the cross-border population, the circle is the floating population, and the triangle is the resident population.
Therefore, in the embodiment, the mobile phone signaling data is combined with the basic attribute information of the mobile phone user to mine and extract the relevant characteristics of the people traveling, and the statistical fields of the number of days of occurrence, the number of days of occurrence of a workplace, the number of days of occurrence of a residential place, the number of days of occurrence of a workplace in a working period, the number of days of occurrence of a residential place in a residential period, and the like are provided.
Feature selection is performed by calculating the total distance entropy between the whole samples and then sorting the features by importance degree by using a backward elimination method.
Based on the screened features, a k-means clustering method is utilized to perform clustering analysis on the mobile phone signaling data, and clustering clusters are divided. And identifying the crowd type of each cluster by combining the distribution condition of each crowd in the corresponding characteristics. Compared with the prior art, the method can more fully mine the information in the mobile phone signaling data, and analyze the category attributes of the crowd from the global perspective by utilizing a machine learning method. The method not only reduces the dependence and the requirement on the prior experience knowledge, improves the applicability of the method, but also can avoid the subjectivity brought by a rule discrimination method.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.
Furthermore, it should be understood that although the present description refers to embodiments, not every embodiment may contain only a single embodiment, and such description is for clarity only, and those skilled in the art should integrate the description, and the embodiments may be combined as appropriate to form other embodiments understood by those skilled in the art.
Claims (4)
1. A crowd type identification method based on mobile phone signaling data is characterized in that:
the method comprises the following steps:
s1, acquiring mobile phone signaling data and corresponding mobile phone user basic attribute information in a research area and a research time period, namely acquiring a data set; the mobile phone signaling data comprises a mobile phone unique identification code, a timestamp, an event type, a base station number, a base station longitude and latitude and a number attribution; the basic attribute information of the mobile phone user comprises age, gender, number attribution, household registration location and a mobile phone unique identification code;
s2, extracting the relevant characteristics of the people in the trip based on the mobile phone signaling data;
s3, forming a feature set by the crowd travel related features extracted in the step S2 and the basic attribute information of the mobile phone users, and screening out an optimal feature subset by using a distance entropy method;
s4, according to the optimal characteristic subset selected in the step S3, carrying out crowd division by using an unsupervised k-means cluster analysis method;
s41, randomly selecting k samples from the sample set as an initial mean vector [ mu ]1,μ2,...,μk}；
S42, traversing the sample set, calculating the Euclidean distance between each sample x and each mean vector, and calculating the Euclidean distance according to the mean vector mu closest to the sample xiDetermining its cluster mark Ci；
S43, after all samples are divided, recalculating the mean vector, wherein the calculation formula is as follows:
s44, repeating the step S42 until the current mean vector is not updated;
s45, output cluster division C ═ C1,C2,...,Ck}；
And S5, carrying out crowd type identification on each cluster according to the characteristic distribution condition of each crowd.
2. The method for crowd type recognition based on mobile phone signaling data as claimed in claim 1, wherein: the crowd trip related feature extraction based on the mobile phone signaling data in the step S2 includes the following steps:
s21, traversing the data set, and respectively counting the number of days of the mobile phone user in the research time range;
s22, determining working time intervals and residence time intervals according to prior experience knowledge and by combining actual conditions of research areas, and identifying places of employment, namely residence and working places, according to the residence time of each residence in the intervals of the mobile phone user; if no stay satisfying the condition exists, recording the place as 0;
and S23, respectively counting the days of the mobile phone users in the residence places, the days of the mobile phone users in the workplace, the days of the mobile phone users in the residence places and the days of the mobile phone users in the workplace based on the identified places of employment.
3. The method for crowd type recognition based on mobile phone signaling data as claimed in claim 1, wherein: the step S3 includes the steps of:
s31, traversing the data set, and calculating Euclidean distance between each sample, sample xiAnd xjHas an Euclidean distance D betweenijThe calculation formula is as follows:
s32, calculating similarity measure between samples and normalizing to be between 0 and 1Sample continuous variable xiAnd xjSimilarity between SijThe calculation formula is as follows:
the similarity calculation formula among the discrete variables is as follows, wherein M is the number of discrete variables:
s33, calculating the distance entropy between each sample, sample xiAnd xjEntropy E of the distance betweenijThe calculation formula is as follows:
Eij＝-SijlogSij-(1-Sij)*log(1-Sij)
s34, calculating the distance entropy of the sample population, wherein the calculation formula is as follows, wherein N represents the number of samples:
s35, calculating the total distance entropy of removing one feature each time by adopting a backward elimination method for the feature set, wherein the feature removed corresponding to the minimum-valued total distance entropy is the least important feature, and then removing the feature from the feature set and putting the feature into a new feature set; and repeating the process for one iteration until all the features are transferred into the new feature set, and reversely ordering the new feature set to obtain the importance ranking of the features.
4. The method for crowd type recognition based on mobile phone signaling data as claimed in claim 1, wherein: step S5 includes
S51, carrying out crowd detailed classification according to the crowd activity characteristic difference and the research precision requirement;
s52, determining approximate spatial distribution of each characteristic of different types of people;
and S53, judging the crowd type of each cluster by combining the crowd characteristic distribution.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910467120.5A CN110245981B (en) | 2019-05-31 | 2019-05-31 | Crowd type identification method based on mobile phone signaling data |
| PCT/CN2020/090065 WO2020238631A1 (en) | 2019-05-31 | 2020-05-13 | Population type recognition method based on mobile phone signaling data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910467120.5A CN110245981B (en) | 2019-05-31 | 2019-05-31 | Crowd type identification method based on mobile phone signaling data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110245981A CN110245981A (en) | 2019-09-17 |
| CN110245981B true CN110245981B (en) | 2021-10-01 |
Family
ID=67885707
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910467120.5A Active CN110245981B (en) | 2019-05-31 | 2019-05-31 | Crowd type identification method based on mobile phone signaling data |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN110245981B (en) |
| WO (1) | WO2020238631A1 (en) |
Families Citing this family (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110245981B (en) * | 2019-05-31 | 2021-10-01 | 南京瑞栖智能交通技术产业研究院有限公司 | Crowd type identification method based on mobile phone signaling data |
| CN110751046B (en) * | 2019-09-19 | 2022-05-17 | 深圳云天励飞技术有限公司 | User identification method and device, readable storage medium and terminal equipment |
| CN111385753B (en) * | 2019-10-24 | 2022-01-04 | 南京瑞栖智能交通技术产业研究院有限公司 | Medical facility accessibility evaluation method based on mobile phone signaling data |
| CN111026738A (en) * | 2019-11-08 | 2020-04-17 | 福建新大陆软件工程有限公司 | Regional population monitoring method and system, electronic equipment and storage medium |
| CN111737605A (en) * | 2020-07-09 | 2020-10-02 | 南京瑞栖智能交通技术产业研究院有限公司 | Travel purpose identification method and device based on mobile phone signaling data |
| CN112070265A (en) * | 2020-07-24 | 2020-12-11 | 北京大学 | Method and device for predicting state of family owning children |
| CN112165686B (en) * | 2020-08-27 | 2021-09-03 | 同济大学 | City visitor identification method, device and storage medium based on mobile phone signaling data |
| CN112512032B (en) * | 2020-11-16 | 2021-09-14 | 南京瑞栖智能交通技术产业研究院有限公司 | Mobile phone signaling data-based external trip crowd identification method |
| CN112633108B (en) * | 2020-11-20 | 2023-04-07 | 移通科技(杭州)有限公司 | People flow monitoring and analyzing system and method based on data management |
| CN112711576B (en) * | 2020-12-11 | 2023-03-10 | 上海城市交通设计院有限公司 | Method for identifying inter-city travel modes of high-speed rail and airplane with mobile phone signaling data |
| CN112906790B (en) * | 2021-02-20 | 2023-08-18 | 国网江苏省电力有限公司营销服务中心 | Solitary old man identification method and system based on electricity consumption data |
| CN113011924A (en) * | 2021-03-24 | 2021-06-22 | 华南理工大学 | Method, system, equipment and medium for identifying logistics distribution and service system of farmer market |
| CN114363823B (en) * | 2021-05-26 | 2023-09-19 | 科大国创云网科技有限公司 | Population density monitoring method and system based on MR (magnetic resonance) resident places and building outlines |
| CN113722878B (en) * | 2021-07-16 | 2022-11-01 | 东南大学 | Simulation-oriented traffic demand determination method based on identity perception data |
| CN113657235A (en) * | 2021-08-10 | 2021-11-16 | 太原理工大学 | Road network traffic jam mode identification method based on image processing |
| CN113918838B (en) * | 2021-11-12 | 2024-04-12 | 合众新能源汽车股份有限公司 | Target crowd identification method, system and readable medium based on stay data |
| CN113935881A (en) * | 2021-12-16 | 2022-01-14 | 北京融信数联科技有限公司 | Population structure analysis method and system based on big data and readable storage medium |
| CN114302333A (en) * | 2021-12-27 | 2022-04-08 | 中国电信股份有限公司 | User identification method and device, electronic equipment and medium |
| CN114219379B (en) * | 2022-02-22 | 2022-05-24 | 北京融信数联科技有限公司 | Resource matching evaluation method and system suitable for community service circle |
| CN114422973B (en) * | 2022-03-30 | 2022-06-28 | 北京融信数联科技有限公司 | Internet taxi appointment driver intelligent identification method and system based on big data and readable storage medium |
| CN115002680B (en) * | 2022-07-28 | 2022-12-27 | 北京融信数联科技有限公司 | Crowd occupation type obtaining method and system based on mobile phone signaling and storage medium |
| CN115409434B (en) * | 2022-11-02 | 2023-03-24 | 北京融信数联科技有限公司 | Regional demographic method, system and storage medium based on signaling big data |
| CN115730763A (en) * | 2022-11-11 | 2023-03-03 | 中山大学 | Method and device for calculating accessibility of facility in workday based on terminal signaling data |
| CN116193369B (en) * | 2022-12-13 | 2023-10-27 | 南京大学 | Resident population generation rate estimation method based on mobile phone signaling data |
| CN116233759B (en) * | 2023-04-25 | 2023-07-28 | 北京城建交通设计研究院有限公司 | Resident travel track investigation method and system |
| CN116992267B (en) * | 2023-09-28 | 2024-01-23 | 北京融信数联科技有限公司 | Regional population gender identification method and system based on signaling data |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105657666A (en) * | 2016-03-31 | 2016-06-08 | 东南大学 | Commercial employee group residence recognition method based on mobile phone positioning data |
| CN105721467A (en) * | 2016-02-16 | 2016-06-29 | 上海交通大学 | Social network Sybil group detection method |
| CN106096631A (en) * | 2016-06-02 | 2016-11-09 | 上海世脉信息科技有限公司 | A kind of recurrent population's Classification and Identification based on the big data of mobile phone analyze method |
| CN107977673A (en) * | 2017-11-15 | 2018-05-01 | 上海世脉信息科技有限公司 | A kind of economically active population's recognition methods based on big data |
| CN109461106A (en) * | 2018-10-11 | 2019-03-12 | 浙江公共安全技术研究院有限公司 | A kind of multidimensional information perception processing method |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104699717B (en) * | 2013-12-10 | 2019-01-18 | 中国银联股份有限公司 | Data digging method |
| CN104123466B (en) * | 2014-07-24 | 2017-07-07 | 中国软件与技术服务股份有限公司 | A kind of big data Study on Trend method for early warning and system based on normal mode |
| CN104156403B (en) * | 2014-07-24 | 2017-08-11 | 中国软件与技术服务股份有限公司 | A kind of big data normal mode extracting method and system based on cluster |
| CN107133318B (en) * | 2017-05-03 | 2021-06-15 | 北京市交通信息中心 | Population identification method based on mobile phone signaling data |
| CN107358265A (en) * | 2017-07-17 | 2017-11-17 | 南京华苏科技有限公司 | External permanent resident population's recognition methods based on spectral clustering |
| CN110245981B (en) * | 2019-05-31 | 2021-10-01 | 南京瑞栖智能交通技术产业研究院有限公司 | Crowd type identification method based on mobile phone signaling data |
-
2019
- 2019-05-31 CN CN201910467120.5A patent/CN110245981B/en active Active
-
2020
- 2020-05-13 WO PCT/CN2020/090065 patent/WO2020238631A1/en active Application Filing
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105721467A (en) * | 2016-02-16 | 2016-06-29 | 上海交通大学 | Social network Sybil group detection method |
| CN105657666A (en) * | 2016-03-31 | 2016-06-08 | 东南大学 | Commercial employee group residence recognition method based on mobile phone positioning data |
| CN105657666B (en) * | 2016-03-31 | 2019-04-30 | 东南大学 | A kind of commercial employment crowd residence recognition methods based on mobile phone location data |
| CN106096631A (en) * | 2016-06-02 | 2016-11-09 | 上海世脉信息科技有限公司 | A kind of recurrent population's Classification and Identification based on the big data of mobile phone analyze method |
| CN107977673A (en) * | 2017-11-15 | 2018-05-01 | 上海世脉信息科技有限公司 | A kind of economically active population's recognition methods based on big data |
| CN109461106A (en) * | 2018-10-11 | 2019-03-12 | 浙江公共安全技术研究院有限公司 | A kind of multidimensional information perception processing method |
Non-Patent Citations (2)
| Title |
|---|
| 基于手机信令数据的昆山市职住分布与通勤特征分析;陆振波等;《现代城市研究》;20190331;第50-55页 * |
| 基于距离熵的应急决策层信息融合方法;管青云等;《系统工程理论与实践》;20150131;第216-227页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2020238631A1 (en) | 2020-12-03 |
| CN110245981A (en) | 2019-09-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110245981B (en) | Crowd type identification method based on mobile phone signaling data | |
| CN109034448B (en) | Trajectory prediction method based on vehicle trajectory semantic analysis and deep belief network | |
| CN103218435B (en) | Method and system for clustering Chinese text data | |
| CN106600960A (en) | Traffic travel origin and destination identification method based on space-time clustering analysis algorithm | |
| CN107977734B (en) | Prediction method based on mobile Markov model under space-time big data | |
| CN111813835B (en) | Public activity center recognition system based on mobile phone signaling and POI data | |
| CN106650273A (en) | Behavior prediction method and device | |
| CN105045858A (en) | Voting based taxi passenger-carrying point recommendation method | |
| Chang et al. | Understanding user’s travel behavior and city region functions from station-free shared bike usage data | |
| CN104573130A (en) | Entity resolution method based on group calculation and entity resolution device based on group calculation | |
| CN112668803B (en) | Automobile service chain enterprise shop-opening and site-selecting method based on LightGBM model | |
| CN112000755B (en) | Regional travel corridor identification method based on mobile phone signaling data | |
| Chen et al. | An analysis of movement patterns between zones using taxi GPS data | |
| CN114661393A (en) | Urban clustering effect visual analysis method based on floating population data feature clustering | |
| CN113158084B (en) | Method, device, computer equipment and storage medium for processing movement track data | |
| CN113240209A (en) | Urban industry cluster development path prediction method based on graph neural network | |
| Xu et al. | MM-UrbanFAC: Urban functional area classification model based on multimodal machine learning | |
| Reyes et al. | Dynamic grouping of vehicle trajectories | |
| CN116013084A (en) | Traffic management and control scene determining method and device, electronic equipment and storage medium | |
| Zhou et al. | Refined taxi demand prediction with ST-Vec | |
| Yang et al. | Mobility pattern identification based on mobile phone data | |
| CN114611622A (en) | Method for identifying cross-city commuting crowd by utilizing mobile phone data | |
| Xia et al. | Predicting human mobility using sina weibo check-in data | |
| Cai et al. | Application of Data Mining Techniques on Tourist Expenses in Malaysia | |
| CN115965137B (en) | Specific object relevance prediction method, system, terminal and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |