WO2015152820A1 - Dispositifs d'identification et procédés d'identification - Google Patents

Dispositifs d'identification et procédés d'identification Download PDF

Info

Publication number
WO2015152820A1
WO2015152820A1 PCT/SG2015/000104 SG2015000104W WO2015152820A1 WO 2015152820 A1 WO2015152820 A1 WO 2015152820A1 SG 2015000104 W SG2015000104 W SG 2015000104W WO 2015152820 A1 WO2015152820 A1 WO 2015152820A1
Authority
WO
WIPO (PCT)
Prior art keywords
trajectories
trajectory
subset
identification
time
Prior art date
Application number
PCT/SG2015/000104
Other languages
English (en)
Inventor
Mingqiang XUE
Huayu WU
Wee Siong NG
Original Assignee
Agency For Science, Technology And Research
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Agency For Science, Technology And Research filed Critical Agency For Science, Technology And Research
Publication of WO2015152820A1 publication Critical patent/WO2015152820A1/fr

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02—Services making use of location information
    • H04W4/029—Location-based management or tracking services
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/535—Tracking the activity of the user

Definitions

  • Embodiments relate generally to identification devices and identification methods.
  • an identification device may be provided.
  • the identification device may include: an input circuit configured to receive a querying trajectory corresponding to a person and a set of trajectories; a subset identification circuit configured to identify a subset of the set of trajectories based on aligning the querying trajectory and the set of trajectories; and an output circuit configured to output the subset as a set of candidates for trajectories corresponding to the person.
  • an identification method may be provided.
  • the identification method may include: receiving a querying trajectory corresponding to a person and a set of trajectories; identifying a subset of the set of trajectories based on aligning the querying trajectory and the set of trajectories; and outputting the subset as a set of candidates for trajectories corresponding to the person.
  • FIG. 1A shows an identification device according to various embodiments
  • FIG. IB shows a flow diagram illustrating an identification method according to various embodiments
  • FIG. 2 shows an illustration of an aligned trajectory, a self-segment and a mutual- segment according to various embodiments. Description
  • the identification device as described in this description may include a memory which is for example used in the processing carried out in the identification device.
  • a memory used in the embodiments may be a volatile memory, for example a DRAM (Dynamic Random Access Memory) or a non-volatile memory, for example a PROM (Programmable Read Only Memory), an EPROM (Erasable PROM), EEPROM (Electrically Erasable PROM), or a flash memory, e.g., a floating gate memory, a charge trapping memoiy, an MRAM (Magnetoresistive Random Access Memory) or a PCRAM (Phase Change Random Access Memory).
  • DRAM Dynamic Random Access Memory
  • PROM Programmable Read Only Memory
  • EPROM Erasable PROM
  • EEPROM Electrical Erasable PROM
  • flash memory e.g., a floating gate memory, a charge trapping memoiy, an MRAM (Magnetoresistive Random Access Memory) or a PCRAM (Phase Change Random Access
  • a “circuit” may be understood as any kind of a logic implementing entity, which may be special potpose circuitry or a processor executing software stored in a memory, firmware, or any combination thereof.
  • a “circuit” may be a hard-wired logic circuit or a programmable logic circuit such as a programmable processor, e.g. a microprocessor (e.g. a Complex Instruction Set Computer (CISC) processor or a Reduced Instruction Set Computer (RISC) processor).
  • a “circuit” may also be a processor executing software, e.g. any kind of computer program, e.g. a computer program using a virtual machine code such as e.g. Java. Any other kind of implementation of the respective functions which will be described in more detail below may also be understood as a "circuit” in accordance with an alternative embodiment.
  • Telco (telecommunication) companies maintain CDR (call detail record) data about every user, including where and when he/she makes a call; Land Transport Authority (LTA) knows the time and original-destination of each EZ-Link card (which is smart card for public transport fare payment in Singapore) user; banks and credit card companies record all the transactions for each customers about the time and location he/she makes the payment; social network service providers keep the check-in information of users, and so on and so forth.
  • LTA Land Transport Authority
  • the Ministry of Health may be keen to control the spread of serious infectious diseases, such as SARS (Severe acute respiratory syndrome). If a particular taxi driver is infected, the ministry may desire to know who are the passengers that were served by this taxi in the last two days, so that isolation and monitoring operations can be taken onto those passengers.
  • the taxi company records the trajectories of the taxi and knows the places and times that the taxi picked up and dropped off a passenger, however, it " cannot know the identity of each passenger.
  • the telco company maintains the CDR data that traces every customer even if he/she is not making a call (the cell phone may keep communicating with the cell tower (or a base station) all the time). If it is possible to match the trajectory of a passenger in the taxi to the CDR trajectory of a particular phone number, the passenger's phone number and probably his/her identity may be known as most telco companies require real-name registration for the mobile service.
  • linking up two or more trajectories of a person can help to gain richer insight of the person's movement. Furthermore, linking up anonymous trajectories with real name trajectories may help to re-identify the person with the anonymous movements.
  • devices and methods may be provided to link different trajectories that belong to the same person. Different from other data integration technologies, which focus on matching two or more data sources with the same content textually or semantically, the devices and methods according to various embodiments may take a probabilistic model to discover the two trajectories that belong to the same person, even though the two trajectories do not share any common data points. Devices and methods according to various embodiments may be used by public and private organizations that have access to the multiple trajectory data sources, to solve different urban problems as demonstrated above, or to provide better social services.
  • devices and methods for fuzzy linking of trajectories datasets for identity discovery may be provided.
  • FIG. 1A shows an identification device 100 according to various embodiments.
  • the identification device 100 may include an input circuit 102 configured to receive (for example read from a storage medium, for example a memory of the identification device 100) a querying trajectory coiTesponding to a person and a set of trajectories.
  • the identification device 100 may further include a subset identification circuit 104 configured to identify a subset of the set of trajectories based on aligning the querying trajectory and the set of trajectories.
  • the identification device 100 may further include an output circuit 106 configured to output the subset as a set of candidates for trajectories coiTesponding to the person.
  • the input circuit 102, the subset identification circuit 104, and the output circuit 106 may be coupled with each other, like indicated by lines 108, for example electrically coupled, for example using a line or a cable, and/ or mechanically coupled.
  • the identification device may align a querying trajectory and another trajectory, and may determine whether the other trajectory is a candidate for a trajectory belonging to the same person like the querying trajectory.
  • each trajectory may include or may be location information and time infonnation.
  • the location infonnation and time information may be ordered according to the time information.
  • aligning two trajectories may include or may be ordering combined location information of two trajectories according to the time information of the two trajectories.
  • the subset identification circuit 104 may be configured to identify a self segment including infonnation of one trajectory, in the aligned trajectory.
  • the subset identification circuit 104 may be configured to identify a mutual segment including infonnation of two different trajectories in the aligned trajectory.
  • the subset identification circuit 104 may be configured to identify for a mutual segment a velocity which is necessary to move from a first location at a first time con-esponding to a first trajectory of the two different trajectories to a second location at a second time corresponding to a second trajectory of the two different trajectories.
  • the subset identification circuit 104 may be configured to determine a compatibility of the aligned trajectory based on the detennined velocity.
  • the subset identification circuit 104 may be configured to identify the subset based on a rejection model. [0026] According to various embodiments, the subset identification circuit 104 may be configured to identify the subset based on an acceptance model.
  • FIG. IB shows a flow diagram 1 10 illustrating an identification method according to various embodiments.
  • a querying trajectory corresponding to a person and a set of trajectories may be received.
  • a subset of the set of trajectories may be identified based on aligning the querying trajectory and at least one trajectory of the set of trajectories.
  • the subset may be output as a set of candidates for trajectories corresponding to the person.
  • each trajectory may include or may be location information and time information.
  • the location information and time information may be ordered according to the time information.
  • aligning two trajectories may include or may be ordering combined location information of two trajectories according to the time information of the two trajectories.
  • the identification method may further include identifying a self segment including information of one trajectory in the aligned trajectory.
  • the identification method may further include identifying a mutual segment including infoimation of two different trajectories in the aligned trajectory.
  • the identification method may further include identifying for a mutual segment a velocity which is necessary to move from a first location at a first time coiTesponding to a first trajectory of the two different trajectories to a second location at a second time coiTesponding to a second trajectory of the two different trajectories.
  • the identification method may further include determining a compatibility of the aligned trajectory based on the determined velocity.
  • the identification method may further include identifying the subset based on a rejection model.
  • the identification method may further include identifying the subset based on an acceptance model.
  • P db ⁇ P 1( ... , P
  • Pj ⁇ and Q db ⁇ Qi, - , Q
  • ⁇ be two different trajectory databases, where each of Pj G is a trajectory of a particular individual in the coiTesponding database.
  • the location information in each record is represented by longitude and latitude.
  • the timestamp of a particular time-location record pj k (resp. qj k ) is denoted as Pi'. t (resp. qj k . t).
  • the location-time records in a trajectory are sorted in time-order, i.e. Pi a . t ⁇ pi b . t if a ⁇ b.
  • three assistance functions may be defined.
  • id(P) is a function for returning the identity of the person whom the trajectory belongs to;
  • dist(p, q) is a function that returns the geographical distance between the location in p and the location in q;
  • timediff(p, q) is a function that returns the absolute difference between the timestamp inpand the timestamp in q.
  • the problem definition for fuzzy linking may be as follows: Given P, a queiying trajectory from a trajectory database P d , and Qd b another non-trial sized trajectory database which contains a trajectory generated by the user id(P), identify a subset of trajectories Q db p Q db where
  • Qdb l > suc h that there is a trajectory Q G Q db p where id(P) id(Q).
  • a small subset of the trajectory database is to be identified, so that the small subset includes a trajectory which belongs to the same person to which the querying trajectory belongs.
  • the above may be the goal of the method. Depending on the quality of the input data, the goal may or may not be realized. If a goal is not realized, the returned subset of trajectories may not include the target trajectory.
  • FIG. 2 shows an illustration 200 of an aligned trajectory 204, a self-segment 206 (which may be a segment which includes only records from one trajectoiy database) and a. mutual-segment 208 (which may be a segment which includes records from two different trajectory databases) according to various embodiments.
  • Time is shown to increase from left to right in FIG. 2, like illustrated by arrow 202.
  • an aligned trajectory W PQ that is formed by aligning P and Q is shown.
  • an aligned trajectory can also be viewed as a sequence of segments, i.e. pairs of(w', w l+ 1 ) where i € [1,
  • a self- segment is a segment (w 1 , w 1+ 1 ) where both w 1 and w l+ 1 are simultaneously from P or Q
  • a mutual-segment is a segment where one of w 1 and w' + 1 is from P and the other is from Q.
  • a self-segment 206 by the pair (q 1 , q 2 ) and a mutual-segment 208 by the pair (p 3 , q 3 ) are labelled.
  • a self- segment is always a segment from the same trajectoiy, but a mutual-segment is a segment that is composed records from two different trajectories.
  • mutual-segments contain useful information for determining if two trajectories belong to the same person and such information may be used for building the rejection-model and acceptance-model.
  • align(P, Q) may be denoted as a function that returns the aligned trajectoiy W PQ from P and Q.
  • a set of statistics may be built that describe the patterns that exist for pairs of trajectories from the same person.
  • the reason for having the keyword "rejection" in the model's name may be that this model is used for "rejecting" an hypothesis in the a rejection phase of an ( 1( a 2 )- filtering method according to various embodiments.
  • the statistics that the models are based on may be decided (or determined). There are two criteria for selecting appropriate statistics for building the models: 1) available, which requires the statistics can be computed directly from available data, i.e.
  • trajectory databases P ⁇ b and Q ⁇ jb the trajectory databases P ⁇ b and Q ⁇ jb ; and 2) discriminative, which requires the rejection-model and acceptance-model to be highly distinguishable by their sets of statistics. Having these criteria, a set of statistics may be chosen for building the models based on mutual-segments compatibility, whose definition may be like described in the following.
  • the definition of compatibility of a mutual segment may be as follows: For a mutual segment (w', w 1+ 1 ) from an aligned trajectory
  • WpQ align(P, Q), (w 1 , w ,+ 1 ) is compatible if and only Vmax '
  • V max is a predefined threshold for the maximum allowed moving speed on the ground. Otherwise, (w ⁇ w 1+ 1 ) is incompatible.
  • the compatibility of a mutual segment is an evaluation for a mutual segment based on whether a person can travel from the location in w' to the location in w 1+ 1 at a reasonable fast speed. For example, the speed of any vehicle is unlikely to be beyond 140 kph (in other words: km/ h) within a town, therefore if the distance between w 1 and w ,+ 1 is 70 km and the time difference is 10 minutes, it is extremely unlikely for one to be able to travel from the location in w 1 to the location in w ,+ 1 using 10 minutes, and hence this mutual segment is incompatible.
  • a naive reasoning is that the incompatible mutual segments are evidence for id(P) ⁇ id(Q).
  • the reasoning may be incorrect, as due to GPS (Global Positioning System) inaccuracy, software or hardware error, in real life data a time-location record does not accurately reflect the exact location of a person at a particular time.
  • GPS Global Positioning System
  • W PQ may still contain incompatible mutual segments.
  • the length of the time-unit may be a user defined parameter, such as half one, one, or two minutes.
  • s r 2 is the probability for a mutual segment of 2 minutes time long (after rounding) to be incompatible, where the mutual segment may be from a trajectory aligned from two trajectories of one person.
  • M r always contains a finite number of statistics. The reason is that, given enough long time, it is always possible for one to travel from one place to another place, meaning mutual segments up to certain time-units long are always compatible.
  • an estimation M r may be learnt from the trajectories in P ⁇ b and Q ⁇ b, by extracting samples from them and feed them to a classification method according to various embodiments.
  • the rejection-model is a set of statistics for mutual segments of trajectories, each of which is aligned from a pair of trajectories of the same person, such mutual segments may build this model according to various embodiments.
  • each of the trajectory in the databases Pdband Q db may be treated as an already-aligned trajectory from a pair of trajectories of the same person and treat each of the segment in the trajectory as a mutual segment.
  • the reason why segments in an individual trajectory may be used as mutual segments for building the M r according to various embodiments may be that both of them are essentially due to the mobility of one person, and their compatibility are only determined by the inherent inaccuracy in location and time information. Therefore, the statistics chosen for the M r also satisfy the available requirement. In the following, it may be assumed that the term rejection-model refers to the estimation M r .
  • the pseudo-code for building M r is shown in Table 1.
  • the acceptance-mode may be given its name for it is used for "accepting" a hypothesis in the a2 -acceptance phase of the (a 1( 2 ) -filtering method according to various embodiments.
  • the set of statistics for building the acceptance-model may be the same like for the rejection-model, except that it may be based on mutual segments of trajectories where each is aligned from two trajectories of different persons.
  • each s a ' (where s a ' > 0) is the probability for mutual segments of i time-units long (after rounding to the nearest integer) to be incompatible in trajectories, where each of the trajectories is aligned from two trajectories of different persons.
  • the acceptance-model may be refeired to as M a .
  • pairs of trajectories of different persons may be used and they may be aligned.
  • the pairs may be obtained by selecting a trajectory from P £ P d b (or Q G Q db ) and pairing it with another trajectory P' G P d b where P ⁇ ?' ( or Q' G Q db where Q ⁇ Q').
  • This approach may be very reasonable as usually a user rarely has more than one trajectory in the same database.
  • the method for building the acceptance-model is shown in Table 2.
  • the core of the method may be how it performs unqualified candidates filtering in the c ⁇ -rejection and 2 -acceptance phases. The details of the two phases are described in more detail below.
  • each P in the a x -rejection phase, each P may be compared against with each trajectory Q € Qdb P - If Q is rejected as a matching, it is removed from Qdb P > otherwise, it is kept and subject to the second phase filtering in the 2 -acceptance phase.
  • the criteria for rejecting Q as a matching is based on how likely, measured by the level of confidence, P and Q are from two different persons.
  • the confidence level is high, i.e. higher than a predefined threshold, Q is rejected as a matching.
  • the confidence in believing that P and Q are from two different persons may be evaluated based on statistical testing with the use of rejection-model.
  • the reasoning of the statistical testing may be as follows: assuming P and Q are indeed from the same person, then the compatibilities of the mutual segments in W P Q, i.e. the alignment of P and Q, should be statistically compliant with the rejection model M R . On the other hand, if it is found that such compliance does not exist, it rather may be believed that the assumption is wrong.
  • the statistical testing described above may require to provide a measurement to the compliance between W P Q and M R .
  • One possible testing statistic for the compliance may be the number of incompatible mutual segments in W P Q . Specifically, under the null hypothesis, each mutual segment in W P Q has a probability to be compatible whose value can be referred from M R based on the time length of the mutual segment.
  • a Poisson-Binomial distribution may be used to describe the distribution of the number of incompatible mutual segments.
  • the Poisson-Binomial distribution is a distribution for the sum of independent Bernoulli trials with not necessarily the same success probabilities.
  • the success probabilities of the " Bernoulli trials may be parameterized by a sequence (p 1( p 2> ... , p n ) where pj is the success probability of the i-th Bernoulli trial and n is the number of trials.
  • the number of Bernoulli trials corresponds to the number of mutual segments in W P Q and the sequence of success probability corresponds (s ⁇ 1 , s r lz , ...
  • K is in Poisson-Binomial distribution parameterized by (s ⁇ 1 , s r lz , ... , s r ln )
  • K is not in Poisson-Binomial distribution parameterized by (s 1 , s r ' 2 , ... , s r ln ) [0063]
  • the null hypothesis H 0 is true and the p-value may be computed for the actual number of incompatible mutual segments in W P Q based on the following distribution function of K:
  • the a ! -rejection phase is effective in pruning pairs of trajectories that are from the different persons, however, not all pairs of trajectories from different persons may be pruned in the 3 ⁇ 4 -rejection phase.
  • the second phase of testing may be provided for the pairs of trajectories that passed the ⁇ ' testing in ax-rejection, which may be referred to as the a 2 -acceptance phase.
  • the purpose may be to accept those pairs of trajectories that passed the testing as trajectories from the same person.
  • the principle for thea 2 -acceptance may be the same as the oCi-rejection except that the testing is based on the acceptance model M a and slightly different hypotheses:
  • H' 0 K is in Poisson-Binomial distribution parameterized by (s 1 , s a ' 2 , ... , s a ln )
  • s ' is the probability for i-th mutual segment whose time length is l j to be incompatible based on M a .
  • the null hypothesis H' 0 may be rejected if the p-value of actual number of incompatible mutual segments in W P Q is smaller than a 2 .
  • the p-value used above may be a term from statistics which may be used for determining the significance of a result.
  • the a x and a 2 may be significance values in statistical test, which may take small values, such as 0.001.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Computer Hardware Design (AREA)
  • General Engineering & Computer Science (AREA)
  • Position Fixing By Use Of Radio Waves (AREA)

Abstract

Selon divers modes de réalisation, la présente invention concerne un dispositif d'identification. Le dispositif d'identification peut comprendre : un circuit d'entrée configuré pour recevoir une trajectoire d'interrogation correspondant à une personne et un ensemble de trajectoires ; un circuit d'identification de sous-ensembles configuré pour identifier un sous-ensemble de l'ensemble de trajectoires sur la base de l'alignement de la trajectoire d'interrogation et d'au moins une trajectoire de l'ensemble de trajectoires ; et un circuit de sortie configuré pour délivrer en sortie le sous-ensemble sous la forme d'un ensemble de candidats pour des trajectoires correspondant à la personne.
PCT/SG2015/000104 2014-03-31 2015-03-31 Dispositifs d'identification et procédés d'identification WO2015152820A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
SG10201401124X 2014-03-31
SG10201401124X 2014-03-31

Publications (1)

Publication Number Publication Date
WO2015152820A1 true WO2015152820A1 (fr) 2015-10-08

Family

ID=54240967

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/SG2015/000104 WO2015152820A1 (fr) 2014-03-31 2015-03-31 Dispositifs d'identification et procédés d'identification

Country Status (1)

Country Link
WO (1) WO2015152820A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114021191A (zh) * 2021-11-05 2022-02-08 江苏安泰信息科技发展有限公司 一种安全生产信息化敏感数据管理方法及系统

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014039241A2 (fr) * 2012-09-05 2014-03-13 Motorola Solutions, Inc. Systèmes et procédés d'analyse et de suivi utilisant des identificateurs en liaison radio de dispositifs mobiles
WO2014039240A1 (fr) * 2012-09-05 2014-03-13 Motorola Solutions, Inc. Procédé et appareil pour identifier un suspect par l'intermédiaire de multiples identités de dispositif mises en corrélation

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014039241A2 (fr) * 2012-09-05 2014-03-13 Motorola Solutions, Inc. Systèmes et procédés d'analyse et de suivi utilisant des identificateurs en liaison radio de dispositifs mobiles
WO2014039240A1 (fr) * 2012-09-05 2014-03-13 Motorola Solutions, Inc. Procédé et appareil pour identifier un suspect par l'intermédiaire de multiples identités de dispositif mises en corrélation

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
CLARKE R.: "PERSON LOCATION AND PERSON TRACKING - TECHNOLOGIES, RISKS AND POLICY IMPLICATIONS", INFORMATION TECHNOLOGY AND PEOPLE, vol. 14, no. 2, 2001, pages 206 - 231 *
MICHAEL K. ET AL.: "LOCATION AND TRACKING OF MOBILE DEVICES: UBERVEILLANCE STALKS THE STREETS", COMPUTER LAW AND SECURITY REVIEW, vol. 29, 2013, pages 216 - 228, XP028539673 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114021191A (zh) * 2021-11-05 2022-02-08 江苏安泰信息科技发展有限公司 一种安全生产信息化敏感数据管理方法及系统

Similar Documents

Publication Publication Date Title
US8924433B2 (en) Methods for geotemporal fingerprinting
CN105404860A (zh) 用于管理走失人员信息的方法和设备
CN107689991B (zh) 信息推送方法和装置、服务器
CN110363076A (zh) 人员信息关联方法、装置及终端设备
CN105894089A (zh) 一种征信模型的建立方法、征信确定方法及对应装置
CN109492122A (zh) 商家信息的获取方法、装置、终端及计算机可读存储介质
CN109902644A (zh) 人脸识别方法、装置、设备和计算机可读介质
KR20070093203A (ko) 전자태그를 이용한 위치정보 관리시스템 및 방법
CN110059849A (zh) 路径确定方法和装置
CN104426933A (zh) 一种筛选增强现实内容的方法、装置及系统
CN105426862A (zh) 基于rfid、定位技术与视频技术的分析方法及其系统
Pei et al. FCN-Attention: A deep learning UWB NLOS/LOS classification algorithm using fully convolution neural network with self-attention mechanism
Ghorpade et al. An integrated stop-mode detection algorithm for real world smartphone-based travel survey
Chen et al. An analysis of movement patterns between zones using taxi GPS data
CN112925899B (zh) 排序模型建立方法、案件线索推荐方法、装置及介质
CN111367956B (zh) 数据统计方法及装置
CN109902129B (zh) 基于大数据分析的保险代理人归类方法及相关设备
CN115002697A (zh) 待核查用户的接触用户识别方法、装置、设备及存储介质
CN111008864B (zh) 一种营销码与商户之间的铺设关系的确定方法、装置及电子设备
US10921132B2 (en) Method to determine search area for a person of interest based upon past route information from searchers
WO2015152820A1 (fr) Dispositifs d'identification et procédés d'identification
CN112235714A (zh) 基于人工智能的poi定位方法、装置、计算机设备及介质
CN116051065A (zh) 考勤数据生成方法、装置、计算机设备及可读存储介质
CN112385180A (zh) 基于交易时间戳将身份和易于获得的个人标识符信息进行匹配的系统和方法
US20220312150A1 (en) Estimation system, estimation method, and information storage medium

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15774293

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase
122 Ep: pct application non-entry in european phase

Ref document number: 15774293

Country of ref document: EP

Kind code of ref document: A1