CN112040414A - Similar track calculation method and device and electronic equipment - Google Patents
Similar track calculation method and device and electronic equipment Download PDFInfo
- Publication number
- CN112040414A CN112040414A CN202010782921.3A CN202010782921A CN112040414A CN 112040414 A CN112040414 A CN 112040414A CN 202010782921 A CN202010782921 A CN 202010782921A CN 112040414 A CN112040414 A CN 112040414A
- Authority
- CN
- China
- Prior art keywords
- signaling
- track
- user
- tracks
- time window
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/20—Services signaling; Auxiliary data signalling, i.e. transmitting data via a non-traffic channel
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Remote Sensing (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Telephonic Communication Services (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A method and a device for calculating similar tracks, electronic equipment and a machine-readable storage medium are disclosed. In the method, all signaling in a specified target time period is obtained from a butted signaling library, and a signaling set to be searched is generated; performing data processing on each signaling in the search signaling set based on a preset time window to generate signaling tracks respectively corresponding to each user, and constructing a signaling track set based on the generated signaling tracks; and executing multi-stage screening calculation on the signaling track set to determine second signaling tracks of other users similar to the first signaling track of the target user, thereby greatly reducing the signaling number and the signaling track number required by similar track calculation and improving the calculation efficiency of similar tracks.
Description
Technical Field
One or more embodiments of the present application relate to the field of computer application technologies, and in particular, to a method and an apparatus for calculating a similar trajectory, an electronic device, and a machine-readable storage medium.
Background
Communication operators have a large number of users, and each user frequently interacts with a base station of the operator through signaling in the process of using the mobile terminal every day, so that more massive signaling is generated. Typically, operators in a city have a number of users on the order of millions or even tens of millions, and the signaling generated daily can be on the order of billions. The signaling can help the operator to carry out more in-depth analysis on the user, form some signaling tracks and provide some valuable data support for organizations such as government public safety departments and the like. However, the data volume of the signaling is too large, and an effective processing mechanism is lacked, so that the value maximization of the user signaling track cannot be realized.
Disclosure of Invention
The application provides a similar track calculation method, which comprises the following steps:
acquiring all the signaling in a specified target time period from a butted signaling library, and generating a signaling set to be searched;
performing data processing on each signaling in the search signaling set based on a preset time window to generate signaling tracks respectively corresponding to each user, and constructing a signaling track set based on the generated signaling tracks;
and performing multi-level screening calculation on the signaling track set to determine second signaling tracks of other users similar to the first signaling track of the target user.
Optionally, the signaling at least includes a user identifier, a reporting time, and a terminal location; the user identification is used for uniquely identifying the user identity, the terminal position represents the geographical position of the terminal carried by the user, and the reporting time represents the corresponding time when the terminal reports the terminal position.
Optionally, the time window is a time period divided into a plurality of time periods with the same length and numbered sequentially;
the processing of the data of each signaling in the search signaling set based on the preset time window to generate the signaling track corresponding to each user respectively includes:
based on the user identification in the signaling, grouping the signaling in the search signaling set according to the user to obtain grouped signaling corresponding to each user;
based on the reporting time in the signaling, carrying out alignment mapping on the grouped signaling respectively corresponding to each user based on a time window to obtain a plurality of signaling aligned and mapped with the time window respectively corresponding to each user;
the terminal position of each signaling in the plurality of signaling after the time window corresponding to each user is aligned and mapped is connected in series according to the serial number sequence of the time window, and a terminal position sequence corresponding to each user is generated;
and generating signaling tracks respectively corresponding to the users based on the generated terminal position sequences respectively corresponding to the users.
Optionally, when multiple signaling with the same user identifier exist in the same time window, the method further includes:
and calculating the central point position corresponding to the terminal position in the multiple signaling, and taking the central point position as the terminal position of the multiple signaling corresponding to the same user identifier in the same time window.
Optionally, the generating, based on the generating of the terminal position sequence corresponding to each user, a signaling track corresponding to each user includes:
respectively generating corresponding hash positions for each terminal position in the terminal position sequence respectively corresponding to each user;
and constructing hash position sequences respectively corresponding to the users based on the generated hash positions, and taking the hash position sequences respectively corresponding to the users as signaling tracks respectively corresponding to the users.
Optionally, the hash position is a GeoHash position;
the performing multi-level screening calculation on the signaling track set to determine a second signaling track of other users similar to the first signaling track of the target user includes:
acquiring a first signaling track of a target user, and calculating and generating a public hash prefix corresponding to a plurality of GeoHash positions included in the first signaling track; the public hash prefix represents a maximum position range corresponding to the first signaling track;
screening out signaling tracks of other users, the GeoHash positions of which are matched with the public Hash prefix, from the signaling track set to generate a track set of other users similar to the first signaling track;
and aligning the signaling tracks according to the time windows based on the time window numbers in the appointed time range, and further screening out second signaling tracks of other users similar to the first signaling track from the track set of other users.
Optionally, aligning the signaling tracks according to the time window numbers in the specified time range, and further filtering out a second signaling track of another user similar to the first signaling track from the track set of another user, including:
respectively unfolding the first signaling track and each signaling track in the track set of other users based on the time window numbers in the specified time range, and respectively aligning the unfolded first signaling track and each signaling track in a time window mode based on the same time window numbers;
calculating whether the matching lengths of the GeoHash positions in the first signaling tracks in the same time window and the same prefixes of the GeoHash positions in the signaling tracks are larger than a preset threshold value or not according to the first signaling tracks and the signaling tracks after the time windows are aligned;
and when the calculated ratio of the number of the time windows larger than the preset threshold value to the total number of the time windows corresponding to the first signaling track reaches a preset first ratio and the non-repetition rate of the GeoHash positions corresponding to the time windows larger than the preset threshold value reaches a preset second ratio, determining the signaling tracks reaching the preset first ratio and the preset second ratio in the track set of other users as second signaling tracks of other users similar to the first signaling tracks.
Optionally, the preset first proportion and the preset second proportion are obtained by inputting a signaling trajectory obtained by random sampling as a negative sample and a similar trajectory labeled by artificial verification as a positive sample into a preset machine learning model to perform model training.
The present application further provides a similar trajectory calculation apparatus, the apparatus comprising:
the first generation module acquires all the signaling in a specified target time period from the butted signaling library and generates a signaling set to be searched;
the second generation module is used for performing data processing on each signaling in the search signaling set based on a preset time window, generating signaling tracks respectively corresponding to each user, and constructing a signaling track set based on the generated signaling tracks;
and the screening calculation module is used for executing multi-stage screening calculation on the signaling track set and determining second signaling tracks of other users similar to the first signaling track of the target user.
Optionally, the signaling at least includes a user identifier, a reporting time, and a terminal location; the user identification is used for uniquely identifying the user identity, the terminal position represents the geographical position of the terminal carried by the user, and the reporting time represents the corresponding time when the terminal reports the terminal position.
Optionally, the time window is a time period divided into a plurality of time periods with the same length and numbered sequentially;
the second generation module further:
based on the user identification in the signaling, grouping the signaling in the search signaling set according to the user to obtain grouped signaling corresponding to each user;
based on the reporting time in the signaling, carrying out alignment mapping on the grouped signaling respectively corresponding to each user based on a time window to obtain a plurality of signaling aligned and mapped with the time window respectively corresponding to each user;
the terminal position of each signaling in the plurality of signaling after the time window corresponding to each user is aligned and mapped is connected in series according to the serial number sequence of the time window, and a terminal position sequence corresponding to each user is generated;
and generating signaling tracks respectively corresponding to the users based on the generated terminal position sequences respectively corresponding to the users.
Optionally, when the same time window corresponds to multiple signaling with the same user identifier, the second generating module further:
and calculating the central point position corresponding to the terminal position in the multiple signaling, and taking the central point position as the terminal position of the multiple signaling corresponding to the same user identifier in the same time window.
Optionally, in the process of generating the signaling tracks corresponding to the respective users based on generating the terminal position sequences corresponding to the respective users, the second generating module further:
respectively generating corresponding hash positions for each terminal position in the terminal position sequence respectively corresponding to each user;
and constructing hash position sequences respectively corresponding to the users based on the generated hash positions, and taking the hash position sequences respectively corresponding to the users as signaling tracks respectively corresponding to the users.
Optionally, the hash position is a GeoHash position;
the screening calculation module further:
acquiring a first signaling track of a target user, and calculating and generating a public hash prefix corresponding to a plurality of GeoHash positions included in the first signaling track; the public hash prefix represents a maximum position range corresponding to the first signaling track;
screening out signaling tracks of other users, the GeoHash positions of which are matched with the public Hash prefix, from the signaling track set to generate a track set of other users similar to the first signaling track;
and aligning the signaling tracks according to the time windows based on the time window numbers in the appointed time range, and further screening out second signaling tracks of other users similar to the first signaling track from the track set of other users.
Optionally, in the process of aligning the signaling tracks according to the time windows based on the time window numbers in the specified time range, further screening out a second signaling track of another user similar to the first signaling track from the track set of another user, where the screening calculation module further:
respectively unfolding the first signaling track and each signaling track in the track set of other users based on the time window numbers in the specified time range, and respectively aligning the unfolded first signaling track and each signaling track in a time window mode based on the same time window numbers;
calculating whether the matching lengths of the GeoHash positions in the first signaling tracks in the same time window and the same prefixes of the GeoHash positions in the signaling tracks are larger than a preset threshold value or not according to the first signaling tracks and the signaling tracks after the time windows are aligned;
and when the calculated ratio of the number of the time windows larger than the preset threshold value to the total number of the time windows corresponding to the first signaling track reaches a preset first ratio and the non-repetition rate of the GeoHash positions corresponding to the time windows larger than the preset threshold value reaches a preset second ratio, determining the signaling tracks reaching the preset first ratio and the preset second ratio in the track set of other users as second signaling tracks of other users similar to the first signaling tracks.
Optionally, the preset first proportion and the preset second proportion are obtained by inputting a signaling trajectory obtained by random sampling as a negative sample and a similar trajectory labeled by artificial verification as a positive sample into a preset machine learning model to perform model training.
The application also provides an electronic device, which comprises a communication interface, a processor, a memory and a bus, wherein the communication interface, the processor and the memory are mutually connected through the bus;
the memory stores machine-readable instructions, and the processor executes the method by calling the machine-readable instructions.
The present application also provides a machine-readable storage medium having stored thereon machine-readable instructions which, when invoked and executed by a processor, perform the method described above.
Through the embodiment, all the signaling in the appointed target time period is obtained from the butted signaling library, and a signaling set to be searched is generated; performing data processing on each signaling in the search signaling set based on a preset time window to generate signaling tracks respectively corresponding to each user, and constructing a signaling track set based on the generated signaling tracks; and executing multi-stage screening calculation on the signaling track set to determine second signaling tracks of other users similar to the first signaling track of the target user, thereby greatly reducing the signaling number and the signaling track number required by similar track calculation and improving the calculation efficiency of similar tracks.
Drawings
FIG. 1 is a flow chart of a method for calculating a similar trajectory provided by an exemplary embodiment;
FIG. 2 is a hardware block diagram of an electronic device provided by an exemplary embodiment;
FIG. 3 is a block diagram of a similar trajectory calculation device provided by an exemplary embodiment.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in this application and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.
It is to be understood that although the terms first, second, third, etc. may be used herein to describe various information, such information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the present application. The word "if" as used herein may be interpreted as "at … …" or "when … …" or "in response to a determination", depending on the context.
In order to make those skilled in the art better understand the technical solution in the embodiment of the present disclosure, the following briefly describes the related art of similar trajectory calculation related to the embodiment of the present disclosure.
Generally, in some scenarios, due to the massive user base and the high frequency of signaling reported by the terminal, in the process of calculating data of a user similar trajectory by a big data system, the signaling required to be calculated is too huge, so that the data calculation amount of the big data system is increased, and the time for calculating the user similar trajectory is too long.
Based on this, the present application aims to provide a technical solution for performing multi-level screening calculation on massive signaling in a signaling library and determining similar trajectories of target users within an acceptable time range.
When the method is realized, the big data system acquires all the signaling in the appointed target time period from the butted signaling library, and generates a signaling set to be searched.
Further, data processing is performed on each signaling in the search signaling set based on a preset time window, signaling tracks respectively corresponding to each user are generated, and a signaling track set is constructed based on the generated signaling tracks.
Further, multi-level screening calculation is performed on the signaling track set, and second signaling tracks of other users similar to the first signaling track of the target user are determined.
In the scheme, all signaling in a specified target time period is acquired from a butted signaling library, and a signaling set to be searched is generated; performing data processing on each signaling in the search signaling set based on a preset time window to generate signaling tracks respectively corresponding to each user, and constructing a signaling track set based on the generated signaling tracks; and executing multi-stage screening calculation on the signaling track set to determine second signaling tracks of other users similar to the first signaling track of the target user, thereby greatly reducing the signaling number and the signaling track number required by similar track calculation and improving the calculation efficiency of similar tracks.
The present application is described below with reference to specific embodiments and specific application scenarios.
Referring to fig. 1, fig. 1 is a flowchart of a similar trajectory calculation method according to an embodiment of the present application, where the method is applied to a big data system, the big data system includes a plurality of big data calculation nodes, and the method performs the following steps:
and 102, acquiring all the signaling in the appointed target time period from the butted signaling library, and generating a signaling set to be searched.
And 104, performing data processing on each signaling in the search signaling set based on a preset time window to generate signaling tracks respectively corresponding to each user, and constructing a signaling track set based on the generated signaling tracks.
And 106, performing multi-level screening calculation on the signaling track set to determine second signaling tracks of other users similar to the first signaling track of the target user.
In this specification, the big data system refers to a machine cluster that performs distributed data based on any big data computing framework.
For example, in practical applications, the big data system may be a machine cluster based on a Hadoop MapReduce offline computing framework, a machine cluster based on an Apache Spark near real-time computing framework, a machine cluster based on an Apache Kafka Streams online real-time computing framework, or a machine cluster based on an HPC high-performance computing framework.
In this specification, the big data system includes a plurality of big data computing nodes for performing data computing processing and a management node for performing cluster management; each big data computing node is pre-allocated with a corresponding computing node identifier, the plurality of big data computing nodes can be respectively registered with the management node, and the management node can allocate, deploy and schedule data computing tasks for the plurality of big data computing nodes.
It should be noted that the big data system may be deployed in a public cloud or a private cloud of an iaas (infrastructure as a service) architecture, and the big data computing node resources may be dynamically increased or decreased in the public cloud or the private cloud, that is, the data computing capability of the big data system may be rapidly improved by horizontal expansion of the dynamic big data computing node resources.
In this specification, the terminal may include any type of terminal that is connected to any carrier data system based on any wired or wireless communication method.
For example, in practical applications, the terminal may be a mobile phone, a platform, a portable device, etc. that is accessed to a mobile/unicom/telecom and other third-party operator data systems through a wireless communication mode based on 2G/3G/4G/5G and Wifi or through a wired network communication mode.
In this specification, the signaling refers to signaling sent in a communication process between the terminal and an access operator data system;
the signaling at least comprises a user identification, a reporting time and a terminal position; the unique user identifier carries the user identity of the user of the terminal, the terminal position represents the geographical position information of the terminal, and the reporting time represents the corresponding time when the terminal reports the terminal position.
For example, in practical application, the user identifier may be any one or a combination of an SN serial number, MAC address information, a mobile phone number of the user, and a user identification number of the terminal. The terminal position may specifically be longitude and latitude information representing geographic position information where the terminal is located. The reporting time may be a time representing a time corresponding to the time when the terminal reports the terminal position, and the time may be a time including year, month, day, hour, minute and second, or a time storing the number of seconds elapsed from 1/1970 as a "Unix time epoch" of a 32-bit/64-bit integer, and a specific time expression manner of the reporting time is not particularly limited in this description.
In this specification, the signaling database refers to a signaling database in an operator data system to which the terminal accesses, the signaling database being used to store the signaling reported by the terminal.
For example, the signaling database may specifically include a signaling database based on a relational database framework or a non-relational database framework.
In this specification, the terminal may report the signaling to the operator data system to which the terminal is accessed at regular time according to a preset time, and the operator data system to which the terminal is accessed may store and record the signaling reported by the terminals to the signaling library.
For example, the terminal may periodically report the signaling to the operator data system to which the terminal is accessed according to a preset time of 30 seconds, and the operator data system to which the terminal is accessed may store and record the signaling of billions of levels reported by millions of terminals respectively into the signaling library.
In this specification, the specified target time interval refers to a time range parameter related to similar trajectory calculation, which is issued to the big data system by a user using the big data system to perform similar trajectory calculation through a UI interface or a command line provided by the big data system.
For example, in practical applications, the specified target time period may specifically include a time interval of signaling required to perform similar trajectory calculation; such as: [ T1, T2 ]; specific formats of T1 and T2 are not specifically limited in this specification, and for example, T1 and T2 may be formats based on year/month/day/hour/minute/second, and may also be formats based on year/month/week/day.
Of course, in practical applications, the specified target time period may also be specifically a specified time range from the current time, such as: the last week, the last three days, etc.
In this specification, the big data system may obtain all the signaling in the specified target time period from the docked signaling library, and generate a set of signaling to be searched.
For example, taking the specified target time period as the last two weeks as an example, the big data system may obtain all the signaling of the reporting time of the signaling in the last two weeks from billions of signaling respectively reported by millions of terminals stored in the signaling library, and generate a set of signaling to be searched, for example: the set of signaling to be searched may specifically include all the signaling reported by all the terminals in the last two weeks.
In this specification, the time window is a preset time period divided into a plurality of time segments with the same length and numbered sequentially.
For example, taking the preset time period as a day as an example, the day may be divided into several time windows with the same length, for example, in a time period with the time window being 5 minutes, the day may be divided into 288 time windows with the length of 5 minutes, and the 288 time windows correspond to time windows which may be numbered as 0, 1, 2,. and 287 in sequence.
The length of the time window is not particularly limited in this specification.
In this specification, after the signaling set to be searched is generated, the big data system performs data processing on each signaling in the searching signaling set based on the preset time window, generates signaling tracks corresponding to each user, and constructs a signaling track set based on the generated signaling tracks.
For example, the big data system performs data processing on each signaling in the search signaling set based on a time window with a length of 5 minutes, generates signaling tracks corresponding to each user, and constructs a signaling track set based on the generated signaling tracks.
In an embodiment shown in the present disclosure, in a process of performing data processing on each signaling in the search signaling set based on the preset time window to generate a signaling track corresponding to each user, the big data system groups the signaling in the search signaling set according to the user identifier in the signaling, so as to obtain a group signaling corresponding to each user.
For example, the search signaling set includes 200 ten thousand signaling reported by 10 ten thousand users in the last two weeks, the big data system performs user grouping on the signaling in the 200 ten thousand signaling based on the user identifier in the signaling to obtain a group signaling corresponding to each user in the 10 ten thousand users, and the group signaling corresponding to each user includes several signaling of the user indicated by the user identifier in the signaling.
In this specification, after obtaining the packet signaling corresponding to each user, the big data system may perform alignment mapping on the packet signaling corresponding to each user based on a time window based on a reporting time in the signaling, to obtain a plurality of signaling after the time window alignment mapping corresponding to each user.
Taking the time window as 5 minutes as an example, after the packet signaling respectively corresponding to 10 ten thousand users is obtained, the big data system may perform alignment mapping on the packet signaling respectively corresponding to 10 ten thousand users based on the time window based on the reporting time in the signaling, so as to obtain a plurality of signaling aligned and mapped on the time windows respectively corresponding to 10 ten thousand users. For example, when a signaling corresponding to a user is aligned and mapped based on a time window, if the reporting time in a certain signaling is located in the time period of the time window with the time window number of 0 (00 hours: 00 minutes: 00 seconds-00 hours: 05 minutes: 00 seconds), the signaling after the time window alignment and mapping is obtained by the signaling and is the signaling with the time window number of 0. By analogy, if the reporting time in a certain signaling is within the time period of the time window with the time window number 287 (23 hours: 55 minutes: 00 seconds-00 hours: 00 minutes: 00 seconds), the signaling corresponding to the signaling after the time window alignment mapping is obtained is the signaling with the time window number 287. Similarly, similar processing is performed for packet signaling corresponding to each of the 10 users after user grouping, and details are not repeated here.
It should be noted that, the above description is only exemplified by the time windows and the time window numbers thereof in one day, and in practical applications, in order to distinguish the time window numbers of the time windows respectively corresponding to multiple days, the signaling may be first screened by day by date, and then the signaling in one day after being screened is aligned and mapped by time windows.
In this specification, after obtaining the plurality of signaling after time window alignment mapping corresponding to each user, the big data system may concatenate the terminal position of each of the plurality of signaling after time window alignment mapping corresponding to each user in the order of time window number, and generate the terminal position sequence corresponding to each user.
Continuing with the example above, after obtaining the signaling after time window alignment mapping corresponding to 10 ten thousand users, the big data system may concatenate the terminal position of each signaling in the signaling after time window alignment mapping corresponding to 10 ten thousand users according to the time window numbering sequence, and generate the terminal position sequence corresponding to 10 ten thousand users. Such as: taking a terminal position sequence corresponding to a certain user as an example, the terminal position sequence may specifically include a terminal position sequence obtained by concatenating terminal positions of each signaling in a plurality of signaling after time window alignment mapping for a plurality of days of the user according to a time window numbering sequence.
In an embodiment shown, in the process of generating a terminal location sequence corresponding to a user in a certain day, when multiple signaling with the same user identifier exist in correspondence to the same time window, the big data system may calculate central point location information corresponding to the terminal location in the multiple signaling, and use the central point location information as the terminal location corresponding to the multiple signaling with the same user identifier in the same time window.
For example, in the process of generating a terminal location sequence corresponding to a user in a certain day, when 2 signaling having the same user identifier exist corresponding to a time window with a time window number of 0, the big data system may calculate center point location information corresponding to a terminal location in the 2 signaling (for example, by calculating an average value of latitudes in the terminal locations of the 2 signaling, center point location information corresponding to a terminal location in the 2 signaling) and use the center point location information as a terminal location corresponding to a time window with a time window number of 0 corresponding to a plurality of signaling having the same user identifier.
It should be noted that, by merging multiple signaling of the same user identifier in the same time window in the signaling trace, the data calculation amount of the big data system is reduced and the calculation efficiency of the big data system for performing similar trace calculation is improved.
In this specification, after generating the terminal position sequences corresponding to the respective users, the big data system may generate the signaling trajectories corresponding to the respective users based on the generation of the terminal position sequences corresponding to the respective users.
In an embodiment shown in the present invention, in the process of generating the signaling tracks corresponding to the respective users based on the generation of the terminal position sequences corresponding to the respective users, the big data system respectively generates corresponding hash positions for each terminal position in the terminal position sequences corresponding to the respective users.
For example, taking a user as an example, the terminal position sequence corresponding to the user includes a terminal position sequence constructed by 1 ten thousand terminal positions (for example, longitude and latitude where the terminal reports signaling), and the big data system respectively generates corresponding hash positions for each terminal position in the terminal position sequence corresponding to each user.
In this specification, the hash position refers to a high-dimensional position generated based on the terminal position in the signaling.
In an embodiment shown in the figure, the hash position is a GeoHash position obtained by calculating the latitude and longitude of the terminal position by a GeoHash algorithm. That is, the GeoHash position can encode the latitude and longitude of two dimensions into a character string of one dimension.
In this specification, after generating a hash position corresponding to each terminal position, the big data system constructs a hash position sequence corresponding to each user, and takes the hash position sequence corresponding to each user as a signaling track corresponding to each user.
For example, the big data system constructs a GeoHash position sequence corresponding to 10 ten thousand users, and takes the GeoHash position sequence corresponding to each user as a signaling track corresponding to each user.
In this specification, after performing data processing on each signaling in the search signaling set based on a preset time window and generating a signaling trajectory corresponding to each user, the big data system may construct a signaling trajectory set based on the generated signaling trajectory.
For example, the big data system may generate 10 ten thousand signaling traces corresponding to 10 ten thousand users, and store the generated 10 ten thousand signaling traces in one set.
In this specification, further, the big data system may perform a multi-level filtering calculation on the signaling trajectory set to determine a second signaling trajectory of another user similar to the first signaling trajectory of the target user.
Continuing the example from the above example, the big data system may perform a multi-level filtering computation on a signaling trajectory set including 10 ten thousand signaling trajectories, and determine second signaling trajectories of one or more other users similar to the first signaling trajectory of the target user.
In an embodiment shown, in the process of performing multi-level screening calculation on the signaling track set and determining second signaling tracks of other users similar to the first signaling track of the target user, the big data system may obtain the first signaling track of the target user and calculate to generate a public hash prefix corresponding to a plurality of GeoHash positions included in the first signaling track; and the public hash prefix represents the maximum position range corresponding to the first signaling track.
For example, based on the process of generating the signaling trace of the user described above, the big data system may obtain the signaling trace a of the target user a in the last two weeks, and calculate and generate a public hash prefix corresponding to a plurality of GeoHash positions included in the signaling trace a; the common hash prefix represents a maximum position range corresponding to the signaling track a, for example, the common hash prefix may represent a position range in a rectangle containing the signaling track a (for example, a rectangle formed by an upper left vertex and a lower right vertex of the signaling track a; or, for example, a rectangle surrounded by tangents of boundary points at the leftmost side, the rightmost side, the uppermost side, and the lowermost side in the signaling track a).
In this specification, the big data system further screens out the signaling tracks of other users whose GeoHash positions match the public hash prefixes from the signaling track set, and generates a track set of other users similar to the first signaling track.
Continuing the example from the above example, the big data system screens out signaling tracks of other users whose GeoHash positions match the public hash prefix from a set of 10 ten thousand signaling tracks (which does not include the signaling track a of the target user a), and generates a set of tracks of other users similar to the signaling track a of the target user a, for example, the set of tracks includes 1 ten thousand tracks of other users similar to the signaling track a.
It should be noted that, when the more the same character string portions matched with the public hash prefix are in the GeoHash position, the closer the GeoHash position is to the GeoHash position range represented by the public hash prefix is, the more the GeoHash position is close to the public hash prefix.
In this specification, the big data system further aligns the signaling tracks according to the time windows based on the time window numbers in the specified time range, and further screens out second signaling tracks of other users similar to the first signaling track from the track set of other users.
Continuing with the above example, the big data system aligns the signaling tracks according to the time windows based on the time window numbers in the last two weeks, and further screens out one or more signaling tracks of other users similar to the signaling track a from the track set of other users including 1 ten thousand signaling tracks.
In an embodiment shown in the above, in the process of aligning the signaling tracks according to the time window numbers in the specified time range and further screening out the second signaling tracks of other users similar to the first signaling track from the track set of other users, the big data system may respectively expand each signaling track in the first signaling track and the track set of other users based on the time window numbers in the specified time range, and respectively time-window-align the expanded first signaling track and each signaling track based on the same time window number.
For example, taking a day in the above specified time range as an example, based on 288 time windows numbered 0 to 287 in the last 1 day (a time window numbering a day divided by 5 minutes), the terminal positions of the first position sequence of user a and the second position sequence of other user B are spread out in the 288 time windows respectively and aligned according to the time windows, as shown in table 1 below:
TABLE 1
In an embodiment shown in the present disclosure, in the process of separately expanding each signaling trajectory in the first signaling trajectory and the trajectory set of the other users, if there is no GeoHash position corresponding to a target time window after the expansion of each signaling trajectory in the first signaling trajectory or the trajectory set of the other users, the big data system may perform linear fitting on GeoHash positions corresponding to time windows respectively adjacent to the time window number of the target time window before and after the time window number, and determine a terminal position obtained by the fitting as the terminal position corresponding to the target time window.
Continuing with the example above, referring to table 1, when PA2 or PB2 corresponding to a target time window (for example, a time window with a time window number of 1 in the target time window table 1) after a signaling track a of a user a or a signaling track of another user B is expanded does not exist (for example, a terminal is restarted/turned off/stuck, and a signal is not good, the terminal position of the corresponding time window is null, so that a GeoHash position calculated according to the terminal position is also null), the big data system may perform linear fitting on GeoHash positions respectively corresponding to time windows (time windows with time window numbers of 0 and 2) adjacent to the time window number of the target time window by using the big data system (PA1 and PA3 may obtain PA2 ', PB1 and 3 may obtain PB 2') and determine PA2 'and PB 2' obtained by the linear fitting as the signaling track a of the user a and the geo hash positions corresponding to the target time window corresponding to the signaling track a and the other user B, respectively.
In this specification, after time window alignment is performed on each signaling track in the expanded first signaling track and the track set of the other users based on the same time window number, for the first signaling track and each signaling track after time window alignment, the big data system may calculate whether a difference between a GeoHash position in the first signaling track and a GeoHash position in each signaling track in the same time window is greater than a preset threshold.
Continuing with the above example, referring to table 1, for 288 time windows numbered 0 to 287, the big data system may respectively calculate whether the matching length of the GeoHash position in the signaling track a of the user a and the same prefix of the GeoHash position in the signaling track B of the other user B in the same time window is greater than a preset threshold (e.g. 6).
In this specification, further, when the calculated ratio of the number of the time windows larger than the preset threshold to the total number of the time windows corresponding to the first signaling trajectory reaches a preset first ratio and the non-repetition rate of the GeoHash position corresponding to the time window larger than the preset threshold reaches a preset second ratio, the big data system determines the signaling trajectory reaching the preset first ratio and the preset second ratio in the trajectory set of the other users as the second signaling trajectory of the other users similar to the first signaling trajectory
Continuing the example from the above example, when the calculated ratio of the number of time windows greater than the preset threshold (for example, the matching length of the same prefix of two GeoHash positions is greater than 6) to the total number of time windows corresponding to the signaling trajectory a reaches a preset first ratio (for example, the first ratio is 50%), and the non-repetition rate of the GeoHash positions corresponding to the time windows greater than the preset threshold reaches a preset second ratio (for example, the second ratio is 30%), the big data system determines the signaling trajectory C that reaches the first ratio and the second ratio in the trajectory set of the other user B as the second signaling trajectory of the other user B similar to the signaling trajectory a.
It should be noted that the non-repetition rate of the GeoHash positions refers to the number of the GeoHash positions corresponding to the time window included in one signaling trace that are different, and is a ratio of the total number of all the GeoHash positions in one signaling trace. Such as: taking the example of 5 minutes in a time window in a day, the total number of the GeoHash positions included in the signaling track is 288, and the user stays in one place for some time (for example, at home for night, at business for work, etc.), and the corresponding number of different GeoHash positions is 100, then the non-repetition rate of the GeoHash positions is 100/288 ≈ 35%
In an embodiment shown in the drawing, the preset first proportion and the preset second proportion are obtained by inputting a signaling track obtained by random sampling as a negative sample and a similar track labeled by artificial check as a positive sample into a preset machine learning model to perform model training.
For example, in practical application, the big data system may randomly sample the signaling trajectory from the signaling trajectory set, and use the randomly sampled signaling trajectory as a negative sample; acquiring a track similar to the signaling track A by the manual check label as a positive sample; inputting the obtained positive sample and the negative sample into a preset machine learning model to execute model training; the preset first proportion and the preset second proportion can be used as model parameters to be trained and optimized in the machine learning model.
The machine learning model is not particularly limited in this specification, and may specifically include a supervised machine learning model (for example, a deep neural network model, etc.) and an unsupervised machine learning model.
In this specification, in practical applications, the big data system may deploy the task of similar trajectory calculation described above to each big data computing node for parallel execution, merge and sort the second signaling trajectories of other users determined by each big data computing node and similar to the first signaling trajectory of the target user, and output other user trajectories similar to the first signaling trajectory.
For example, in a suspected contagion screening scenario of an infectious disease patient, the big data system performs merged sorting on the signaling tracks of the contagions determined by each big data computing node to be similar to the first signaling track of the infectious disease patient in a specified time period, outputs the second signaling tracks of all the contagions similar to the first signaling track of the infectious disease patient, and performs priority sorting output based on the similarity degree of the tracks.
In the technical scheme, all signaling in a specified target time period is acquired from a butted signaling library, and a signaling set to be searched is generated; performing data processing on each signaling in the search signaling set based on a preset time window to generate signaling tracks respectively corresponding to each user, and constructing a signaling track set based on the generated signaling tracks; executing multi-stage screening calculation to the signaling track set to determine the second signaling tracks of other users similar to the first signaling track of the target user, and greatly reducing the signaling number and the signaling track number required by similar track calculation, thereby improving the calculation efficiency of similar tracks
For example, in practical applications, similar trajectory calculation needs to be performed for signaling of billions of levels originally, and with the above-described embodiments, signaling for performing similar trajectory calculation may be greatly needed, such as: reducing by four to five orders of magnitude (from billions of levels to hundreds of thousands of levels, or even tens of thousands of levels), and greatly reducing the number of signaling tracks required by similar track calculation, thereby improving the calculation efficiency of similar tracks.
Corresponding to the above method embodiments, the present application also provides embodiments of a similar trajectory calculation apparatus.
Corresponding to the above method embodiments, the present specification also provides an embodiment of a similar trajectory calculation device. The embodiment of the similar trajectory calculation device in the present specification can be applied to electronic equipment. The device embodiments may be implemented by software, or by hardware, or by a combination of hardware and software. Taking a software implementation as an example, as a logical device, the device is formed by reading, by a processor of the electronic device where the device is located, a corresponding computer program instruction in the nonvolatile memory into the memory for operation. From a hardware aspect, as shown in fig. 2, the hardware structure diagram of an electronic device in which a similar trajectory calculation device of this specification is located is shown, except for the processor, the memory, the network interface, and the nonvolatile memory shown in fig. 2, the electronic device in which the device is located in the embodiment may also include other hardware according to the actual function of the electronic device, which is not described again.
FIG. 3 is a block diagram of a similar trajectory calculation device as shown in an exemplary embodiment of the present description.
Referring to fig. 3, the similar trajectory calculation apparatus 30 can be applied to the electronic device shown in fig. 2, and includes:
the first generation module 301 obtains all the signaling in the specified target time period from the docked signaling library, and generates a signaling set to be searched;
a second generating module 302, configured to perform data processing on each signaling in the search signaling set based on a preset time window, generate signaling tracks corresponding to each user, and construct a signaling track set based on the generated signaling tracks;
and the screening calculation module 303 performs multi-level screening calculation on the signaling trajectory set to determine second signaling trajectories of other users similar to the first signaling trajectory of the target user.
In this embodiment, the signaling at least includes a user identifier, a reporting time, and a terminal location; the user identification is used for uniquely identifying the user identity, the terminal position represents the geographical position of the terminal carried by the user, and the reporting time represents the corresponding time when the terminal reports the terminal position.
In this embodiment, the time window is a time period divided into a plurality of time periods with the same length and numbered sequentially;
the second generation module 302 further:
based on the user identification in the signaling, grouping the signaling in the search signaling set according to the user to obtain grouped signaling corresponding to each user;
based on the reporting time in the signaling, carrying out alignment mapping on the grouped signaling respectively corresponding to each user based on a time window to obtain a plurality of signaling aligned and mapped with the time window respectively corresponding to each user;
the terminal position of each signaling in the plurality of signaling after the time window corresponding to each user is aligned and mapped is connected in series according to the serial number sequence of the time window, and a terminal position sequence corresponding to each user is generated;
and generating signaling tracks respectively corresponding to the users based on the generated terminal position sequences respectively corresponding to the users.
In this embodiment, when the same time window corresponds to multiple signaling with the same user identifier, the second generating module 302 further:
and calculating the central point position corresponding to the terminal position in the multiple signaling, and taking the central point position as the terminal position of the multiple signaling corresponding to the same user identifier in the same time window.
In this embodiment, in the process of generating the signaling tracks corresponding to the respective users based on generating the terminal position sequences corresponding to the respective users, the second generating module 302 further:
respectively generating corresponding hash positions for each terminal position in the terminal position sequence respectively corresponding to each user;
and constructing hash position sequences respectively corresponding to the users based on the generated hash positions, and taking the hash position sequences respectively corresponding to the users as signaling tracks respectively corresponding to the users.
In this embodiment, the hash position is a GeoHash position;
the screening calculation module 303 further:
acquiring a first signaling track of a target user, and calculating and generating a public hash prefix corresponding to a plurality of GeoHash positions included in the first signaling track; the public hash prefix represents a maximum position range corresponding to the first signaling track;
screening out signaling tracks of other users, the GeoHash positions of which are matched with the public Hash prefix, from the signaling track set to generate a track set of other users similar to the first signaling track;
and aligning the signaling tracks according to the time windows based on the time window numbers in the appointed time range, and further screening out second signaling tracks of other users similar to the first signaling track from the track set of other users.
In this embodiment, in the process of aligning the signaling tracks according to the time windows based on the time window numbers in the specified time range, and further screening out the second signaling tracks of other users from the track set of other users, which are similar to the first signaling track, the screening calculation module 303 further:
respectively unfolding the first signaling track and each signaling track in the track set of other users based on the time window numbers in the specified time range, and respectively aligning the unfolded first signaling track and each signaling track in a time window mode based on the same time window numbers;
calculating whether the matching lengths of the GeoHash positions in the first signaling tracks in the same time window and the same prefixes of the GeoHash positions in the signaling tracks are larger than a preset threshold value or not according to the first signaling tracks and the signaling tracks after the time windows are aligned;
and when the calculated ratio of the number of the time windows larger than the preset threshold value to the total number of the time windows corresponding to the first signaling track reaches a preset first ratio and the non-repetition rate of the GeoHash positions corresponding to the time windows larger than the preset threshold value reaches a preset second ratio, determining the signaling tracks reaching the preset first ratio and the preset second ratio in the track set of other users as second signaling tracks of other users similar to the first signaling tracks.
In this embodiment, the preset first proportion and the preset second proportion are obtained by inputting a signaling trajectory obtained by random sampling as a negative sample and a similar trajectory labeled by manual checking as a positive sample into a preset machine learning model to perform model training.
The apparatuses, modules or modules illustrated in the above embodiments may be implemented by a computer chip or an entity, or by an article with certain functions. A typical implementation device is a computer, which may take the form of a personal computer, laptop computer, cellular telephone, camera phone, smart phone, personal digital assistant, media player, navigation device, email messaging device, game console, tablet computer, wearable device, or a combination of any of these devices.
Other embodiments of the present disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This specification is intended to cover any variations, uses, or adaptations of the specification following, in general, the principles of the specification and including such departures from the present disclosure as come within known or customary practice within the art to which the specification pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the specification being indicated by the following claims.
It will be understood that the present description is not limited to the precise arrangements described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present description is limited only by the appended claims.
The above description is only a preferred embodiment of the present disclosure, and should not be taken as limiting the present disclosure, and any modifications, equivalents, improvements, etc. made within the spirit and principle of the present disclosure should be included in the scope of the present disclosure.
Claims (11)
1. A method of similar trajectory computation, the method comprising:
acquiring all the signaling in a specified target time period from a butted signaling library, and generating a signaling set to be searched;
performing data processing on each signaling in the search signaling set based on a preset time window to generate signaling tracks respectively corresponding to each user, and constructing a signaling track set based on the generated signaling tracks;
and performing multi-level screening calculation on the signaling track set to determine second signaling tracks of other users similar to the first signaling track of the target user.
2. The method of claim 1, wherein the signaling at least includes a user identifier, a reporting time, and a terminal location; the user identification is used for uniquely identifying the user identity, the terminal position represents the geographical position of the terminal carried by the user, and the reporting time represents the corresponding time when the terminal reports the terminal position.
3. The method of claim 2, wherein the time window is a preset time period divided into a plurality of time segments with the same length and being numbered sequentially;
the processing of the data of each signaling in the search signaling set based on the preset time window to generate the signaling track corresponding to each user respectively includes:
based on the user identification in the signaling, grouping the signaling in the search signaling set according to the user to obtain grouped signaling corresponding to each user;
based on the reporting time in the signaling, carrying out alignment mapping on the grouped signaling respectively corresponding to each user based on a time window to obtain a plurality of signaling aligned and mapped with the time window respectively corresponding to each user;
the terminal position of each signaling in the plurality of signaling after the time window corresponding to each user is aligned and mapped is connected in series according to the serial number sequence of the time window, and a terminal position sequence corresponding to each user is generated;
and generating signaling tracks respectively corresponding to the users based on the generated terminal position sequences respectively corresponding to the users.
4. The method of claim 3, when there are multiple signaling of the same user identifier corresponding to the same time window, further comprising:
and calculating the central point position corresponding to the terminal position in the multiple signaling, and taking the central point position as the terminal position of the multiple signaling corresponding to the same user identifier in the same time window.
5. The method of claim 3, wherein generating the signaling trajectory corresponding to each user based on generating the terminal position sequence corresponding to each user comprises:
respectively generating corresponding hash positions for each terminal position in the terminal position sequence respectively corresponding to each user;
and constructing hash position sequences respectively corresponding to the users based on the generated hash positions, and taking the hash position sequences respectively corresponding to the users as signaling tracks respectively corresponding to the users.
6. The method of claim 5, the hash location being a GeoHash location;
the performing multi-level screening calculation on the signaling track set to determine a second signaling track of other users similar to the first signaling track of the target user includes:
acquiring a first signaling track of a target user, and calculating and generating a public hash prefix corresponding to a plurality of GeoHash positions included in the first signaling track; the public hash prefix represents a maximum position range corresponding to the first signaling track;
screening out signaling tracks of other users, the GeoHash positions of which are matched with the public Hash prefix, from the signaling track set to generate a track set of other users similar to the first signaling track;
and aligning the signaling tracks according to the time windows based on the time window numbers in the appointed time range, and further screening out second signaling tracks of other users similar to the first signaling track from the track set of other users.
7. The method of claim 6, the aligning signaling traces by time window based on time window numbers within the specified time range, further filtering out from the set of traces of other users a second signaling trace of other users similar to the first signaling trace, comprising:
respectively unfolding the first signaling track and each signaling track in the track set of other users based on the time window numbers in the specified time range, and respectively aligning the unfolded first signaling track and each signaling track in a time window mode based on the same time window numbers;
calculating whether the matching lengths of the GeoHash positions in the first signaling tracks in the same time window and the same prefixes of the GeoHash positions in the signaling tracks are larger than a preset threshold value or not according to the first signaling tracks and the signaling tracks after the time windows are aligned;
and when the calculated ratio of the number of the time windows larger than the preset threshold value to the total number of the time windows corresponding to the first signaling track reaches a preset first ratio and the non-repetition rate of the GeoHash positions corresponding to the time windows larger than the preset threshold value reaches a preset second ratio, determining the signaling tracks reaching the preset first ratio and the preset second ratio in the track set of other users as second signaling tracks of other users similar to the first signaling tracks.
8. The method according to claim 7, wherein the preset first ratio and the preset second ratio are obtained by inputting a similar trajectory labeled by manual check as a positive sample to a preset machine learning model to perform model training based on a signaling trajectory obtained by random sampling as a negative sample.
9. A similar trajectory calculation device, the device comprising:
the first generation module acquires all the signaling in a specified target time period from the butted signaling library and generates a signaling set to be searched;
the second generation module is used for performing data processing on each signaling in the search signaling set based on a preset time window, generating signaling tracks respectively corresponding to each user, and constructing a signaling track set based on the generated signaling tracks;
and the screening calculation module is used for executing multi-stage screening calculation on the signaling track set and determining second signaling tracks of other users similar to the first signaling track of the target user.
10. An electronic device comprises a communication interface, a processor, a memory and a bus, wherein the communication interface, the processor and the memory are connected with each other through the bus;
the memory has stored therein machine-readable instructions, the processor executing the method of any of claims 1 to 8 by calling the machine-readable instructions.
11. A machine readable storage medium having stored thereon machine readable instructions which, when invoked and executed by a processor, carry out the method of any of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010782921.3A CN112040414B (en) | 2020-08-06 | 2020-08-06 | Similar track calculation method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010782921.3A CN112040414B (en) | 2020-08-06 | 2020-08-06 | Similar track calculation method and device and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112040414A true CN112040414A (en) | 2020-12-04 |
| CN112040414B CN112040414B (en) | 2023-04-07 |
Family
ID=73582614
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010782921.3A Active CN112040414B (en) | 2020-08-06 | 2020-08-06 | Similar track calculation method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112040414B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105205155A (en) * | 2015-09-25 | 2015-12-30 | 珠海世纪鼎利科技股份有限公司 | Big data criminal accomplice screening system and method |
| US20180032858A1 (en) * | 2015-12-14 | 2018-02-01 | Stats Llc | System and method for predictive sports analytics using clustered multi-agent data |
| CN108595539A (en) * | 2018-04-04 | 2018-09-28 | 烟台海颐软件股份有限公司 | A kind of recognition methods of trace analogical object and system based on big data |
| US20190056423A1 (en) * | 2016-03-25 | 2019-02-21 | Alibaba Group Holding Limited | Adjoint analysis method and apparatus for data |
| CN110263840A (en) * | 2019-06-13 | 2019-09-20 | 东软集团股份有限公司 | A kind of circuit analysis method, apparatus and program product, storage medium |
| CN110443156A (en) * | 2019-07-16 | 2019-11-12 | 浙江大华技术股份有限公司 | Track method for measuring similarity, data processing equipment and storage equipment |
| CN110944296A (en) * | 2019-11-27 | 2020-03-31 | 智慧足迹数据科技有限公司 | Accompanying determination method and device of motion trail and server |
| CN111209261A (en) * | 2020-01-02 | 2020-05-29 | 邑客得（上海）信息技术有限公司 | User travel track extraction method and system based on signaling big data |
| CN111291278A (en) * | 2020-01-16 | 2020-06-16 | 深圳市前海随手数据服务有限公司 | Method and device for calculating track similarity, storage medium and terminal |
| CN111460074A (en) * | 2020-04-09 | 2020-07-28 | 深圳云天励飞技术有限公司 | Trajectory collision analysis method and device and electronic equipment |
-
2020
- 2020-08-06 CN CN202010782921.3A patent/CN112040414B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105205155A (en) * | 2015-09-25 | 2015-12-30 | 珠海世纪鼎利科技股份有限公司 | Big data criminal accomplice screening system and method |
| US20180032858A1 (en) * | 2015-12-14 | 2018-02-01 | Stats Llc | System and method for predictive sports analytics using clustered multi-agent data |
| US20190056423A1 (en) * | 2016-03-25 | 2019-02-21 | Alibaba Group Holding Limited | Adjoint analysis method and apparatus for data |
| CN108595539A (en) * | 2018-04-04 | 2018-09-28 | 烟台海颐软件股份有限公司 | A kind of recognition methods of trace analogical object and system based on big data |
| CN110263840A (en) * | 2019-06-13 | 2019-09-20 | 东软集团股份有限公司 | A kind of circuit analysis method, apparatus and program product, storage medium |
| CN110443156A (en) * | 2019-07-16 | 2019-11-12 | 浙江大华技术股份有限公司 | Track method for measuring similarity, data processing equipment and storage equipment |
| CN110944296A (en) * | 2019-11-27 | 2020-03-31 | 智慧足迹数据科技有限公司 | Accompanying determination method and device of motion trail and server |
| CN111209261A (en) * | 2020-01-02 | 2020-05-29 | 邑客得（上海）信息技术有限公司 | User travel track extraction method and system based on signaling big data |
| CN111291278A (en) * | 2020-01-16 | 2020-06-16 | 深圳市前海随手数据服务有限公司 | Method and device for calculating track similarity, storage medium and terminal |
| CN111460074A (en) * | 2020-04-09 | 2020-07-28 | 深圳云天励飞技术有限公司 | Trajectory collision analysis method and device and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112040414B (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10540611B2 (en) | Scalable complex event processing with probabilistic machine learning models to predict subsequent geolocations | |
| US10877947B2 (en) | Deduplication of metadata for places | |
| US20210374204A1 (en) | Method and apparatus for managing region tag | |
| Chen et al. | Constructing and comparing user mobility profiles | |
| Cao et al. | Automatic user identification method across heterogeneous mobility data sources | |
| CN107590250A (en) | A kind of space-time orbit generation method and device | |
| CN109325118B (en) | Unbalanced sample data preprocessing method and device and computer equipment | |
| Doyle et al. | Population mobility dynamics estimated from mobile telephony data | |
| US9135630B2 (en) | Systems and methods for large-scale link analysis | |
| Ghosh et al. | Modeling of human movement behavioral knowledge from gps traces for categorizing mobile users | |
| Li et al. | Mining periodicity from dynamic and incomplete spatiotemporal data | |
| CN109257383A (en) | A kind of BGP method for detecting abnormality and system | |
| CN109145225B (en) | Data processing method and device | |
| CN111148018A (en) | Method and device for identifying and positioning regional value based on communication data | |
| CN112925899B (en) | Ordering model establishment method, case clue recommendation method, device and medium | |
| CN113837635A (en) | Risk detection processing method, device and equipment | |
| CN112040414B (en) | Similar track calculation method and device and electronic equipment | |
| CN113468226A (en) | Service processing method, device, electronic equipment and storage medium | |
| CN112040413B (en) | User track calculation method and device and electronic equipment | |
| CN106886510A (en) | The method and device that a kind of chart shows | |
| CN106600053B (en) | User attribute prediction system based on space-time trajectory and social network | |
| CN115438719A (en) | Data processing method, device, server and storage medium | |
| CN113254672B (en) | Method, system, equipment and readable storage medium for identifying abnormal account | |
| CN113076451B (en) | Abnormal behavior identification and risk model library establishment method and device and electronic equipment | |
| Zhou et al. | Sedyt: A general framework for multi-step event forecasting via sequence modeling on dynamic entity embeddings |
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 |