CN103218442A - Method and system for life mode analysis based on mobile device sensor data - Google Patents
Method and system for life mode analysis based on mobile device sensor data Download PDFInfo
- Publication number
- CN103218442A CN103218442A CN2013101417262A CN201310141726A CN103218442A CN 103218442 A CN103218442 A CN 103218442A CN 2013101417262 A CN2013101417262 A CN 2013101417262A CN 201310141726 A CN201310141726 A CN 201310141726A CN 103218442 A CN103218442 A CN 103218442A
- Authority
- CN
- China
- Prior art keywords
- data
- carried out
- user
- mobile device
- sensor
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- 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
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a method and a system for life mode analysis based on mobile device sensor data. The method comprises the steps: initial data are collected from each sensor of a mobile device; data preprocessing is conducted on the initial data according to data characteristics and energy consumption characteristics of each sensor and a data sequence is obtained; a resident point sequence is obtained according to a resident point detection mode; cluster analysis is conducted on the resident point sequence so that a site historical sequence is obtained; interest point search is conducted on each piece of data in the site historical sequence and site history is marked; and whether identity of a user is known or not is judged, if the answer is negative, the identity of the user is deduced according to interest point marks. The method and the system for the life mode analysis based on the mobile device sensor data reduce data mining and analyzing cost, improve using flexibility, have transportability and are convenient to use and capable of improving experience of the user.
Description
Technical field
The present invention relates to the data mining technology field, relate in particular to a kind of life pattern analytical approach and system based on the mobile device sensing data.
Background technology
At present, usually all be built-in with various sensors in the mobile device, comprise GPS (Global Positioning System, GPS), accelerometer, gyroscope, microphone, distance perspective should wait, the user can obtain the raw data of these sensor record by the interface that system provides.Exist at present and user's odd-numbered day data are carried out data mining, thereby find the algorithm that the user stops the place based on gps data.Aspect other sensing data analyses, exist the data of obtaining based on wearable sensor device to come the method that user's behavior is inferred at present.Aspect data mining, there is the clustering method based on density such as OPTICS and DBSCAN at present; Aspect the point of interest search, there is more Map Services to open the relevant interface of point of interest search to third party developer both at home and abroad.
Analyze and sum up at present correlative study based on the life data mining of sensing data, nothing more than two big classes: the first kind is in different rooms with sensor stationary distribution of the same race, collect data respectively, the sensing data of each position is gathered the back on time and position two dimensions, carry out analyzing and processing; Second class is that portable sensor moves with the user and the positional information of recording user in real time, on time dimension user position information is analyzed after the combined data.
As shown in Figure 1, basic procedure based on the life data mining of sensor network is: at first collected raw data and be aggregated into database by each sensor that is distributed in diverse geographic location (being generally in each room) and preserve, after having collected enough data, respectively the raw data of each sensor is analyzed according to time series, itself and historical data are analyzed.
As shown in Figure 2, require the user to carry sensor device, by analyzing the feature of each sensing data on time and space, thereby user environment is judged, and then user's life pattern is extracted based on the portable equipment sensing data.
Existing technical scheme mainly contains following shortcoming:
(1) for technical scheme shown in Figure 1:
Need to lay sensor network in advance in fixing room, cost is higher, use scene dumb (not having portability) and do not have versatility; In addition, also need shift to an earlier date being connected between sensors configured and the computing machine, comparatively loaded down with trivial details.
(2) for technical scheme shown in Figure 2:
Require the user to carry extra electronic equipment at any time or wear wearable device, certain hardware cost is also arranged; And portable this province of sensor device can not deal with data, data need be sent to computing machine with Wireless transmission mode and handle, and dirigibility is relatively poor.
Summary of the invention
The objective of the invention is to overcome the deficiencies in the prior art, the invention provides a kind of life pattern analytical approach and system based on the mobile device sensing data, can reduce the cost of data mining and analysis, and the dirigibility of raising use, and has portability, easy to use, improve user's experience property.
In order to address the above problem, the present invention proposes a kind of life pattern analytical approach based on the mobile device sensing data, described method comprises:
Collect raw data from each sensor of mobile device;
Data characteristics and energy consumption characteristics according to each sensor are carried out the data pre-service and are obtained data sequence raw data;
Obtain the dwell point sequence according to the dwell point detection mode;
Described dwell point sequence is carried out cluster analysis, obtain the place historical series;
Each bar data in the historical series of place are carried out the point of interest search, and place history is carried out mark;
Whether the identity of judging the user is known, if not, then infers user identity according to the point of interest mark.
Preferably, described raw data comprises locator data and motion state raw data.
Preferably, described each sensor from mobile device step of collecting raw data comprises: general CoreLocation framework is collected locator data; By the collect motion state raw data of self-acceleration meter, gyroscope, electronic compass of CoreMotion framework.
Preferably, described locator data comprises cell tower triangle locator data, Wi-Fi node database data query and GPS locator data.
Preferably, described data characteristics and the energy consumption characteristics step of raw data being carried out the data pre-service and obtaining data sequence according to each sensor comprises: the locator data of collecting is carried out coordinate transform; Coordinate after the conversion is carried out reverse geocoding; User ID, longitude and latitude, timestamp, motor pattern, reverse geographical coding result and Record ID are write in the database.
Preferably, described data characteristics and the energy consumption characteristics step of raw data being carried out the data pre-service and obtaining data sequence according to each sensor comprises: each sensing data is constantly carried out Filtering Processing; Use second normal form to characterize each data of sensor constantly respectively; Safeguard the formation of a second normal form; Second normal form sequence in the formation is carried out Fourier transform; Calculate its power spectrum; Constitute fingerprint vector from the bigger data of power spectrum intercepting discrimination; On frequency field, carry out similarity relatively with the motor pattern that presets; With the highest title that presets motor pattern of similarity user's current motor pattern is carried out mark.
Correspondingly, the embodiment of the invention also discloses a kind of life pattern analytic system based on the mobile device sensing data, described system comprises:
Collection module is used for collecting raw data from each sensor of mobile device;
Pretreatment module is used for according to the data characteristics and the energy consumption characteristics of each sensor raw data being carried out the data pre-service and being obtained data sequence;
The dwell point processing module is used for obtaining the dwell point sequence according to the dwell point detection mode;
The cluster analysis module is used for described dwell point sequence is carried out cluster analysis, obtains the place historical series;
Mark module is used for each bar data of place historical series are carried out the point of interest search, and place history is carried out mark;
Judge module is used to judge whether user's identity is known, if not, then infers user identity according to the point of interest mark.
Preferably, described collection module is used for general CoreLocation framework collection locator data; By the collect motion state raw data of self-acceleration meter, gyroscope, electronic compass of CoreMotion framework.
Preferably, described pretreatment module also is used for the locator data of collecting is carried out coordinate transform; Coordinate after the conversion is carried out reverse geocoding; User ID, longitude and latitude, timestamp, motor pattern, reverse geographical coding result and Record ID are write in the database.
Preferably, described pretreatment module also is used for each sensing data is constantly carried out Filtering Processing; Use second normal form to characterize each data of sensor constantly respectively; Safeguard the formation of a second normal form; Second normal form sequence in the formation is carried out Fourier transform; Calculate its power spectrum; Constitute fingerprint vector from the bigger data of power spectrum intercepting discrimination; On frequency field, carry out similarity relatively with the motor pattern that presets; With the highest title that presets motor pattern of similarity user's current motor pattern is carried out mark.
The present invention may operate on the smart mobile phone that most of people carry every day, has better flexibility and portability; Owing to need not to lay in advance the fixed sensing network or carry extra wearable device, so the user need not additionally to purchase other hardware devices, data mining and analysis with low cost, and comparatively convenient; Collect raw data from the various sensors that mobile device is built-in, and program run is on the backstage, the user need not carry out any extra configuration operation after entering software, easy and simple to handle, can exempt uninteresting loaded down with trivial details manual input process in the Traditional affair annoyware effectively, make the affairs prompt function intellectuality and the robotization more of customer mobile terminal; In addition, can be effectively information such as user's behavior pattern for living, user identity be made identification and inferred, and then generate personalized affairs prompt scheme at the user automatically, this prompting scheme will be carried out corresponding optimization process at user's behavior track, generate targetedly based on time and location-based prompting, and, and do not need extra configuration along with the change of user's life pattern can be carried out the adjustment of affairs prompt scheme adaptively; And the software that this scheme realizes can utilize the built-in annoyware of the corresponding interface and system mutual mutually, makes the affairs prompt that generates can be synchronized to automatically on other equipment of user, and comparatively convenient, user experience is good.
Description of drawings
In order to be illustrated more clearly in the embodiment of the invention or technical scheme of the prior art, to do to introduce simply to the accompanying drawing of required use in embodiment or the description of the Prior Art below, apparently, accompanying drawing in describing below only is some embodiments of the present invention, for those of ordinary skills, under the prerequisite of not paying creative work, can also obtain other accompanying drawing according to these accompanying drawings.
Fig. 1 is the basic framework synoptic diagram that has now based on the life data mining of sensor network;
Fig. 2 is the basic framework synoptic diagram that has now based on the life data mining of portable equipment sensing data;
The schematic flow sheet based on the life pattern analytical approach of mobile device sensing data of Fig. 3 embodiment of the invention;
Fig. 4 is the basic configuration diagram of the involved software of embodiment of the invention method;
Fig. 5 is the process synoptic diagram of the preprocessing process of locator data in the embodiment of the invention;
Fig. 6 is the process synoptic diagram of the processing procedure of accelerometer data in the embodiment of the invention;
Fig. 7 is the synoptic diagram of second normal form formation in the embodiment of the invention;
Fig. 8 is the synoptic diagram of intercepting finger print information in the embodiment of the invention;
Fig. 9 is a schematic flow sheet of collecting the motor pattern finger print information in the embodiment of the invention;
Figure 10 is a dwell point testing process synoptic diagram in the embodiment of the invention;
Figure 11 is based on the synoptic diagram of central point periphery P OI keyword search in the embodiment of the invention;
Figure 12 is the process synoptic diagram of inferring based on the user identity of central point labeled bracketing in the embodiment of the invention;
Figure 13 is the configuration diagram that generates affairs prompt in the embodiment of the invention automatically;
Figure 14 is that the structure based on the life pattern analytic system of mobile device sensing data of the embodiment of the invention is formed synoptic diagram.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the invention, the technical scheme in the embodiment of the invention is clearly and completely described, obviously, described embodiment only is the present invention's part embodiment, rather than whole embodiment.Based on the embodiment among the present invention, those of ordinary skills belong to the scope of protection of the invention not making the every other embodiment that is obtained under the creative work prerequisite.
Usually all be built-in with various sensors in the mobile device, they have write down a large amount of raw data, remain therefrom to be excavated more valuable information.At present, a large number of users gets used to carrying every day smart mobile phone or panel computer, and its sensing data has write down user's animation and behavior pattern well.In the present invention, the raw data that each sensor is collected from mobile device is started with, and at the characteristics of different sensors, sets up the model of corresponding data processing.At first data are carried out pre-service, carry out data mining then, therefrom extract user's life pattern, and then can utilize the life pattern information of obtaining to provide all kinds of services targetedly the user.
The inventive method and system are applicable to the present overwhelming majority's mobile intelligent terminal, so have universality preferably.Software is the C-S framework, develops client on mobility device, its server end of exploitation on publicly-owned cloud platform.At first, software should wait the sensor acquisition of information by the built-in GPS of mobile device, accelerometer, gyroscope, microphone, distance perspective, is incorporated into range monitoring, based on the modes such as text analyzing of social networks, obtains user profile and uploads onto the server; The information that the server end analysis-by-synthesis is obtained, the information of evolution data collection, constantly near this user's life pattern, progressively can be automatically after the training for user's generation based on the time with based on the reminding service of the personalization in place, and write built-in annoyware automatically.
Below in conjunction with Fig. 3, Fig. 4 the implementation procedure based on the life pattern analytical approach of mobile device sensing data of the embodiment of the invention is elaborated.
Fig. 3 is the life pattern analytical approach based on the mobile device sensing data of the embodiment of the invention, and as shown in Figure 3, this method comprises:
S301 collects raw data from each sensor of mobile device;
S302 carries out the data pre-service and obtains data sequence raw data according to the data characteristics and the energy consumption characteristics of each sensor;
S303 obtains the dwell point sequence according to the dwell point detection mode;
S304 carries out cluster analysis to the dwell point sequence, obtains the place historical series;
S305 carries out the point of interest search to each the bar data in the historical series of place, and place history is carried out mark;
S306 judges whether user's identity is known, if not, then infers user identity according to the point of interest mark, and then carries out point of interest search procedure more accurately once more at user identity.
As shown in Figure 4, be example with the iOS platform, set forth the involved software basic framework of embodiment of the invention method.Server is responsible for the lasting user data of preserving, so that the user uses distinct device to carry out data aggregation easily.The CoreMotion framework is responsible for obtaining the data of user's sensors such as accelerometer, gyroscope and electronic compass; The CoreLocation framework is responsible for obtaining from cell tower, Wi-Fi access point and gps satellite user's position data; The ShareKit framework obtains the text message of user social contact network account transmission and is uploaded to the SAE server and preserves.After the data of collecting q.s, handle and analyze in this locality after the client-requested data download.
In S301, on different intelligent mobile phone platforms, corresponding interface is arranged, be that example is set forth with iOS.Adopt the CoreLocation framework to collect locator data, locator data comprises cell tower triangle locator data, Wi-Fi node database data query and GPS locator data.Use in these three kinds of modes one or more according to different ambient conditions, thereby when guaranteeing degree of precision, also will keep lower energy consumption.Can collect by the CoreMotion framework motion state raw data of sensors such as self-acceleration meter, gyroscope, electronic compass.
In embodiments of the present invention, the preprocessing process of locator data at first carries out coordinate transform to the location raw data of collecting as shown in Figure 5; Then the coordinate after the conversion is carried out reverse geocoding, in order to auxiliary judgment; At last, user ID, longitude and latitude, timestamp, motor pattern, reverse geographical coding result and Record ID are write in the database.Below will be specifically at setting forth in detail wherein.
In concrete the enforcement, the geographic coordinate data that GPS is collected be designated as (latitude, longitude).Because these data are based on the WGS-84 coordinate system, and need carry out the non-linear inclined to one side processing that adds to geographical coordinate information according to State Bureau of Surveying and Mapping, to meet the encryption standard algorithm of state survey office.Coordinate after the conversion is GCJ-02 coordinate system (Guojia Cehui Ju-02Coordinate System), the coordinate after the conversion is designated as (latitude ', longitude '); To the coordinate after the conversion (latitude ', longitude ') carries out reverse geocoding (Reverse Geocoding), thereby obtain abundant more environmental information, such as street, building name etc., be designated as (thoroughfare, building), these environmental informations play auxiliary effect to the analysis of late time data and to the judgement of customer location; And the longitude and latitude after user ID, the conversion, timestamp, motor pattern, reverse geographical coding result together write database together with Record ID.
In addition, the processing procedure of accelerometer data as shown in Figure 6.Earlier each sensing data is constantly carried out Filtering Processing; Use second normal form to characterize each data of sensor constantly then respectively; Dynamically safeguard the formation of a second normal form; Again the second normal form sequence in the formation is carried out Fourier transform; Calculate its power spectrum; Constitute fingerprint vector from the bigger data of power spectrum intercepting discrimination then; And then on frequency field, carry out similarity relatively with the motor pattern that presets; At last, with the highest title that presets motor pattern of similarity user's current motor pattern is carried out mark.
Except that timestamp, the acceleration information of collecting has three dimensions, and being respectively the component of this vector on x, y, z axle is real number all, carries out quadrature and decompose resulting three components and be designated as accX, accY, accZ respectively in the space.Consider in the process of collecting data, can not guarantee that the user carries the sensing of equipment, so in concrete enforcement, consider three dimension equivalent processes.Yet because acceleration of gravity (gravity) straight down, has clear and definite directivity, so consider to eliminate the interference of acceleration of gravity to experimental data by Hi-pass filter (High-Pass Filter) is set.Sampling rate is made as 60Hz, cutoff frequency is made as 5Hz, filtered data are designated as (accX ', accY ', accZ ').
Owing to do not consider the orientation of equipment itself, so adopt the second normal form of data after the calculation of filtered herein, it can reflect the instantaneous characteristics of sensing data effectively, and can reduce data dimension effectively, reduce the complexity of data processing, reduce the calculated amount expense of resource-constrained mobile device as much as possible.The sequence of the second normal form that calculates is designated as (2NF1,2NF2 ..., 2NFn-1,2NFn).Frequency with 60Hz is sampled to accelerometer, constantly deposits the second normal form of data in nearest a second in formation (Queue), Figure 7 shows that second normal form formation synoptic diagram.
When upgrading positional information, the data sequence in the formation is carried out Fast Fourier Transform (FFT), utilizing input signal in algorithm all is the characteristics of real number, computation process is carried out its corresponding simplified, and the result after the Fourier transform is designated as (freq0, freq1,, freqn/2+1).Then the frequency data after the conversion are calculated corresponding power spectrum, and power sequence be (pow0, pow1 ..., pown/2+1).
Consider the factors such as space complexity of algorithm, choose the fingerprint (Fingerprint) of low-frequency plurality of data here, in order to this motor pattern of mark as motor pattern, Fig. 8 is the synoptic diagram of intercepting finger print information, and the fingerprint of note intercepting is (pow0, pow1, pow2, pow3, pow4).
Be illustrated in figure 9 as the basic flow sheet of collecting the motor pattern finger print information.The mode of averaging by long-time collection data, to jump, run, the frequency spectrum feature of walking and common motor pattern (Motion Pattern) such as static calculates and analyzes, the result after handling is preserved as built-in motor pattern.Then with power spectrum as a vector, carry out similarity with the power spectrum of built-in motor pattern and calculate.With the similarity comparative approach that adopts based on Euclidean distance, with the result of calculation stipulations to (0,1] in the interval, computing formula is:
Find out and the immediate built-in motor pattern of current data by the mode of similarity comparison, thereby user's motion state is carried out mark.
In order to reduce the energy consumption of mobile device as much as possible, the filtrator (Distance Filter) of adjusting the distance adopts the mode that dynamically arranges, be that to be moved beyond the position of upgrading with last time be center, assign thresholds (threshold) when be radius regional user's position, trust (delegate) method just can be carried out position renewal.
When each position renewal, the instantaneous velocity when the location-based service of system is obtained current location and upgraded is if the size of this speed, can conclude then that the current motion state of user is low-speed motions such as static or walking less than 2m/s, so be set to 1m apart from filtrator; And when user's movement velocity during greater than 2m/s, the motion state that can conclude the user is for running or taking other vehicles, since under such motion state, can not produce dwell point, the threshold value apart from filtrator can be brought up to 10m, to reduce the energy consumption of equipment.
After obtaining user position update, carry out coordinate transform with regard to request server pair warp and weft degree immediately.Every record comprises the data such as longitude and latitude, timestamp, motor pattern and mark after Record ID, user ID, the conversion.Set up the database that comprises user message table and data logger at server end, in order to storage user's relevant information.Table 1 is pretreated data recording example.
The pretreated data recording example of table 1
Below S303 is further set forth.
In embodiments of the present invention, dwell point (Stay Point) probe algorithm is improved, make its can be effectively in the scene that many days sensing datas of unique user are handled to the user not the sensing data sequence of same date handle once.Figure 10 is a dwell point testing process synoptic diagram.
In the embodiment of the invention, realized being applicable to dwell point (Stay Point) detection algorithm of the many day data of single user, so that handle to many days data are disposable.The geographic coordinate information that is input as of improved dwell point monitoring algorithm is output as dwell point coordinate and corresponding timestamp, the dwell point sequence information is deposited in the local file of terminal.Consideration is in actual conditions, and the user collects the whole day data and starts from usually and get up morning, ends at before the sleep in evening in the shutdown.In original dwell point detects, if the last data point that mobile device collects just the user in the family, so because the user does not exceed the distance threshold in this algorithm, so a last dwell point will be lost.In the testing process of dwell point, the data of same date are not distinguished, and, avoided the disappearance of dwell point data effectively whether existing dwell point to carry out independent judgement in the data sequence at last.Geographic coordinate after the conversion is carried out dwell point detect, and then generate the dwell point sequence, be designated as (sp0, sp1 ..., spn).
Below S304 is described in detail.
Because what consider is the cluster analysis of the sensing data of unique user, do not expect to obtain the hierarchical clustering result, and in the described scene of the embodiment of the invention, the point in the data acquisition might constitute various irregular shapes, also inevitably can comprise some noises in the data.Take all factors into consideration the factor of aspects such as these problems and calculated amount, here adopt DBSCAN cluster (the Density-Based Spatial Clustering of Applications with Noise) algorithm that is fit to these problems of processing that dwell point is carried out cluster analysis, be equivalent to the input data point has been carried out the division of single aspect.
Yet, because this algorithm is same parameter Epsilon and MinimumPoints to each dimension employing, handle so need carry out " normalization ", make the data of different pieces of information scope or even non-commensurate can under identical parameter and standard, carry out cluster analysis effectively the data on the different dimensions.In fact the DBSCAN cluster analysis concentrates all points in fact to carry out a division to data, these points or belong to some bunch that algorithm generates, or be noise just.Output result to the DBSCAN algorithm has carried out further processing, and the mean value that calculates the longitude of the point in each bunch and latitude respectively is as each bunch center point coordinate.To bunch in each point sort according to the timestamp in the record, the mark of each point point of subscript median in array in calculating bunch is as the mark of this bunch central point.
Data point is carried out cluster analysis on three dimensions, these three dimensions are respectively longitude (Latitude), latitude (Longitude) and time of arrival (ArrivingTime).Because temporal information is not a floating point type, has carried out pre-service earlier at this, the defined feature value is to calculate eigenwert the considering as the time dimension numerical values recited of each point timestamp.Obtain the center and the corresponding timestamp information of each cluster after the cluster, deposit it in local file.
In inventive embodiments, (Point Of Interest, POI) information such as mark is inferred user's current behavior by analysis-by-synthesis accelerometer data and point of interest.Infer user's job characteristics according to analyzing the mark that the POI search obtains, and then adopt different strategies when inferring user behavior, thereby improve the accuracy when inferring behavior once more, make that the making prompting that generates automatically is more accurate.
In addition, the specific implementation process of S305 is as follows:
Then be that the nearest POI name of central point of adjusting the distance is referred to as the mark of central point in model, concrete processing procedure is as follows:
At first, be central point with the GCJ-02 coordinate after each conversion, be foundation with information such as its corresponding timestamp and user identity, initiate searching request to server accordingly, and after server end obtains Search Results, with the result asynchronous return to client; Client traversal POI Search Results promptly is referred to as the mark of this position history (Location History) apart from the distance (POI Distance) of respective center point apart from the name of a nearest POI of central point.Figure 11 shows that synoptic diagram based on central point periphery P OI keyword search.
The specific implementation process of S306 is as follows:
In that having been carried out, all central points behind the mark mark of these all central points of user is classified.The user of different identity has different separately points of interest.For example, minor student's point of interest generally includes school, teaching building, library etc.; And the employee's who grows up point of interest generally includes company, market etc.Classify by mark, come this user of mark more to meet which kind of user identity, and then choose the identity information of a maximum classification of mark as this user to customer center point.
After inferring this user's identity, just can utilize subscriber identity information to optimize the lists of keywords of POI search targetedly.In addition, user's identity information also provides important basis for the affairs prompt that customization is provided.As shown in figure 12, Figure 12 infers (Identity Speculation) process synoptic diagram based on the user identity of central point labeled bracketing.
Below in conjunction with the method that the embodiment of the invention provided several potential application scenarioss are described.
A. the generation of reminding based on the automatic processing of when and where
Can set up system architecture shown in Figure 13, soft the carrying out alternately of Reminders that employing EventKit framework and iOS system are built-in.The user identity that obtains according to deduction is in conjunction with the point of interest mark of user at diverse location, for the user formulates corresponding reminded contents; Set up the inference rule storehouse, its distinguishing rule comprises the timestamp information of data point, reverse geographical coded message, user's all kinds of information that obtain with analysis such as occupational information.For example, if user identity is the university student, and it returns the words of passing by the supermarket on the road of dormitory at night, can be when this user arrives near the supermarket, remind him to buy breakfast bread of second day etc. in passing, also can plan, or the prompting of healthy aspect is provided according to the rule degree of user's work and rest user's schedule.The automatic prompting that generates uploads to the Apple server, thereby realizes in user's distinct device automatically synchronously.
In such scene, the logical mode of social networks text analyzing and process monitoring of can also introducing is obtained the more information about user's habits and customs, thereby generates personalized more affairs prompt.Carry out text analyzing by information, thereby obtain more a plurality of people's information of user the user social contact network; Also can carry out data mining, therefrom analyze characteristics and other relevant informations of this area, to promote the accuracy that life pattern is analyzed other users' of this area the social networks text of being sent out.Can by the title of the analysis software that the user uses under different time and place condition, thereby obtain the custom that the user uses software in real time at the process list of background monitoring subscriber equipment.Be on duty such as, user and all can come the important news of every day in the way, can before the user goes to work, open news software automatically so with news software.All can open the mail that the mailbox client is replied the same day at once after going back home every night, so can the current position of supervisory user, after the user gets home evening, just open the mailbox client automatically.
B. the automatic monitoring and the prompting of user health situation
At present a lot of people are in a kind of state that medically is called " inferior health " because habits and customs are bad.This state initial stage does not have tangible illness, thus often ignored by people, and then health status go fromes bad to worse.Can be under the situation that guarantees privacy of user and data security, collect the mechanism of user sensor data in the integrated timing of internal system, by analysis and excavation to sensing data, come the health status of monitor user ' by the behavioural habits of analysis user, thereby the healthalert of personalization is provided for this user.Such as user's little motion, often work overtime or work and rest situation such as irregular when taking place, system can remind the user to take exercises or work and rest regularly in time.
C. the automatic switchover of device profile
The automatic switchover smart mobile phone contextual model, switch between the built-in contextual model of system according to the varying environment of different location.Such as in the library environment that needs peace and quiet, automatically switching to " silent mode ", can exempt interference, then automatically switch to " outdoor pattern " at noisy outdoor environment, avoid can't hear the tinkle of bells.And enter in the car the user, can whether be the driver by further differentiation user, thereby whether decision change contextual model into " vehicle-mounted pattern " automatically, also can generate affairs prompt more targetedly at the driver.
D. good friend's commending system of hitchhiking
Collecting under the situation of a large amount of different user datas, can in the social networks class is used, carry out friend recommendation, getting to know and hitchhike jointly such as the people that can recommend similar route on and off duty based on the life pattern similarity.Because the life pattern of different user has certain similarity, can design all kinds of commending systems based on this, and all kinds of intelligentized services are provided.
E. Ge Xinghua advertisement pushing service
Can be by the sensing data of analysis user, thereby obtain user's habits and customs and personal preference, such as certain user has the custom of body-building venue, golf course or commercial club, can come to formulate more targetedly and push the advertisement of commodity or service according to these information, thereby obtain better to promote effect.
The method of the embodiment of the invention may operate on the smart mobile phone that most of people carry every day, has better flexibility and portability; Owing to need not to lay in advance the fixed sensing network or carry extra wearable device, so the user need not additionally to purchase other hardware devices, data mining and analysis with low cost, and comparatively convenient; In addition, collect raw data from the various sensors that mobile device is built-in, and program run is on the backstage, the user need not carry out any extra configuration operation after entering software, easy and simple to handle, can exempt uninteresting loaded down with trivial details manual input process in the Traditional affair annoyware effectively, make the affairs prompt function intellectuality and the robotization more of customer mobile terminal; This method can be effectively made identification and is inferred information such as user's behavior pattern for living, user identity, and then generate personalized affairs prompt scheme at the user automatically, this prompting scheme will be carried out corresponding optimization process at user's behavior track, generate targetedly based on time and location-based prompting, and, and do not need extra configuration along with the change of user's life pattern can be carried out the adjustment of affairs prompt scheme adaptively; And the software that this scheme realizes can utilize the built-in annoyware of the corresponding interface and system mutual mutually, makes the affairs prompt that generates can be synchronized to automatically on other equipment of user, and comparatively convenient, user experience is good.
In addition, the embodiment of the invention also discloses a kind of life pattern analytic system based on the mobile device sensing data, as shown in figure 14, this system comprises:
Dwell point processing module 3 is used for obtaining the dwell point sequence according to the dwell point detection mode;
Mark module 5 is used for each bar data of place historical series are carried out the point of interest search, and place history is carried out mark;
Wherein, collection module 1 is used for general CoreLocation framework collection locator data; By the collect motion state raw data of self-acceleration meter, gyroscope, electronic compass of CoreMotion framework.
Can describe about Principle of Process among the embodiment referring to the life pattern analytical approach based on the mobile device sensing data of the present invention based on the implementation procedure of each functions of modules of the life pattern analytic system of mobile device sensing data and principle in the embodiment of the invention repeats no more here.
One of ordinary skill in the art will appreciate that all or part of step in the whole bag of tricks of the foregoing description is to instruct relevant hardware to finish by program, this program can be stored in the computer-readable recording medium, storage medium can comprise: ROM (read-only memory) (ROM, Read Only Memory), random access memory (RAM, Random Access Memory), disk or CD etc.
In addition, more than life pattern analytical approach and system based on the mobile device sensing data that the embodiment of the invention provided are described in detail, used specific case herein principle of the present invention and embodiment are set forth, the explanation of above embodiment just is used for helping to understand method of the present invention and core concept thereof; Simultaneously, for one of ordinary skill in the art, according to thought of the present invention, the part that all can change in specific embodiments and applications, in sum, this description should not be construed as limitation of the present invention.
Claims (10)
1. life pattern analytical approach based on the mobile device sensing data is characterized in that described method comprises:
Collect raw data from each sensor of mobile device;
Data characteristics and energy consumption characteristics according to each sensor are carried out the data pre-service and are obtained data sequence raw data;
Obtain the dwell point sequence according to the dwell point detection mode;
Described dwell point sequence is carried out cluster analysis, obtain the place historical series;
Each bar data in the historical series of place are carried out the point of interest search, and place history is carried out mark;
Whether the identity of judging the user is known, if not, then infers user identity according to the point of interest mark.
2. the life pattern analytical approach based on the mobile device sensing data as claimed in claim 1 is characterized in that described raw data comprises locator data and motion state raw data.
3. the life pattern analytical approach based on the mobile device sensing data as claimed in claim 2 is characterized in that, the step that described each sensor from mobile device is collected raw data comprises: general CoreLocation framework is collected locator data; By the collect motion state raw data of self-acceleration meter, gyroscope, electronic compass of CoreMotion framework.
4. as claim 2 or 3 described life pattern analytical approachs, it is characterized in that described locator data comprises cell tower triangle locator data, Wi-Fi node database data query and GPS locator data based on the mobile device sensing data.
5. the life pattern analytical approach based on the mobile device sensing data as claimed in claim 4, it is characterized in that described data characteristics and the energy consumption characteristics step of raw data being carried out the data pre-service and obtaining data sequence according to each sensor comprises: the locator data of collecting is carried out coordinate transform; Coordinate after the conversion is carried out reverse geocoding; User ID, longitude and latitude, timestamp, motor pattern, reverse geographical coding result and Record ID are write in the database.
6. the life pattern analytical approach based on the mobile device sensing data as claimed in claim 4, it is characterized in that described data characteristics and the energy consumption characteristics step of raw data being carried out the data pre-service and obtaining data sequence according to each sensor comprises: each sensing data is constantly carried out Filtering Processing; Use second normal form to characterize each data of sensor constantly respectively; Safeguard the formation of a second normal form; Second normal form sequence in the formation is carried out Fourier transform; Calculate its power spectrum; Constitute fingerprint vector from the bigger data of power spectrum intercepting discrimination; On frequency field, carry out similarity relatively with the motor pattern that presets; With the highest title that presets motor pattern of similarity user's current motor pattern is carried out mark.
7. life pattern analytic system based on the mobile device sensing data is characterized in that described system comprises:
Collection module is used for collecting raw data from each sensor of mobile device;
Pretreatment module is used for according to the data characteristics and the energy consumption characteristics of each sensor raw data being carried out the data pre-service and being obtained data sequence;
The dwell point processing module is used for obtaining the dwell point sequence according to the dwell point detection mode;
The cluster analysis module is used for described dwell point sequence is carried out cluster analysis, obtains the place historical series;
Mark module is used for each bar data of place historical series are carried out the point of interest search, and place history is carried out mark;
Judge module is used to judge whether user's identity is known, if not, then infers user identity according to the point of interest mark.
8. the life pattern analytic system based on the mobile device sensing data as claimed in claim 7 is characterized in that, described collection module is used for general CoreLocation framework and collects locator data; By the collect motion state raw data of self-acceleration meter, gyroscope, electronic compass of CoreMotion framework.
9. the life pattern analytic system based on the mobile device sensing data as claimed in claim 7 is characterized in that, described pretreatment module also is used for the locator data of collecting is carried out coordinate transform; Coordinate after the conversion is carried out reverse geocoding; User ID, longitude and latitude, timestamp, motor pattern, reverse geographical coding result and Record ID are write in the database.
10. the life pattern analytic system based on the mobile device sensing data as claimed in claim 6 is characterized in that, described pretreatment module also is used for each sensing data is constantly carried out Filtering Processing; Use second normal form to characterize each data of sensor constantly respectively; Safeguard the formation of a second normal form; Second normal form sequence in the formation is carried out Fourier transform; Calculate its power spectrum; Constitute fingerprint vector from the bigger data of power spectrum intercepting discrimination; On frequency field, carry out similarity relatively with the motor pattern that presets; With the highest title that presets motor pattern of similarity user's current motor pattern is carried out mark.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013101417262A CN103218442A (en) | 2013-04-22 | 2013-04-22 | Method and system for life mode analysis based on mobile device sensor data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013101417262A CN103218442A (en) | 2013-04-22 | 2013-04-22 | Method and system for life mode analysis based on mobile device sensor data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103218442A true CN103218442A (en) | 2013-07-24 |
Family
ID=48816229
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2013101417262A Pending CN103218442A (en) | 2013-04-22 | 2013-04-22 | Method and system for life mode analysis based on mobile device sensor data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103218442A (en) |
Cited By (36)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103533171A (en) * | 2013-10-21 | 2014-01-22 | 英华达（上海）科技有限公司 | Method and control device for automatic mobile phone profile switching and mobile phone with automatic mobile phone profile switching function |
| CN103543832A (en) * | 2013-10-29 | 2014-01-29 | Tcl集团股份有限公司 | User identification method and device based on unbalanced data |
| CN103714139A (en) * | 2013-12-20 | 2014-04-09 | 华南理工大学 | Parallel data mining method for identifying a mass of mobile client bases |
| CN104077404A (en) * | 2014-07-07 | 2014-10-01 | 西安交通大学 | Online taxpayer identity recognition method based on variable-length system calling sequence birthmarks |
| CN104252527A (en) * | 2014-09-02 | 2014-12-31 | 百度在线网络技术（北京）有限公司 | Method and device for determining resident point information of mobile subscriber |
| CN104615881A (en) * | 2015-01-30 | 2015-05-13 | 南京烽火星空通信发展有限公司 | User normal track analysis method based on movable position application |
| WO2015067119A1 (en) * | 2013-11-07 | 2015-05-14 | 华为技术有限公司 | Method for clustering position points of interest and related device |
| WO2015081480A1 (en) * | 2013-12-03 | 2015-06-11 | 深圳绿拓科技有限公司 | Room occupancy state sensing method and room occupancy state sensing apparatus |
| CN104731795A (en) * | 2013-12-19 | 2015-06-24 | 日本电气株式会社 | Mining apparatus and method of activity patterns of individuals |
| CN105045792A (en) * | 2014-04-30 | 2015-11-11 | 三星电子株式会社 | Apparatus and method for integrated management of data in mobile device, and mobile device |
| CN105307121A (en) * | 2015-10-16 | 2016-02-03 | 上海晶赞科技发展有限公司 | Information processing method and device |
| CN105302900A (en) * | 2014-10-24 | 2016-02-03 | 康复广告株式会社 | Computer implemented method and system for social network service |
| CN105388820A (en) * | 2014-08-25 | 2016-03-09 | 深迪半导体(上海)有限公司 | Intelligent monitoring device and monitoring method thereof, and monitoring system |
| CN105760837A (en) * | 2016-02-18 | 2016-07-13 | 西北工业大学 | Method for identifying group moving pattern based on heterogeneous sensor |
| DE102015212703B3 (en) * | 2015-07-07 | 2016-07-28 | Technische Universität Dresden | Method and apparatus for determining properties of at least one sub-micron structure |
| CN106022810A (en) * | 2015-02-13 | 2016-10-12 | 成都易驾科技有限公司 | Advertisement audience targeting and putting system based on driving behavior of car owner |
| CN106294485A (en) * | 2015-06-05 | 2017-01-04 | 华为技术有限公司 | Determine the method and device in notable place |
| CN106339456A (en) * | 2016-08-26 | 2017-01-18 | 重庆科创职业学院 | Push method based on data mining |
| CN106384120A (en) * | 2016-08-29 | 2017-02-08 | 深圳先进技术研究院 | Mobile phone positioning data based resident activity pattern mining method and device |
| CN106407277A (en) * | 2016-08-26 | 2017-02-15 | 北京车网互联科技有限公司 | Internet of vehicles data-based attribute analysis method for vehicle owner parking point after being clustered |
| CN106708034A (en) * | 2016-11-23 | 2017-05-24 | 河池学院 | Implementation method for controlling walking path of robot on basis of mobile phone |
| CN107171872A (en) * | 2017-07-19 | 2017-09-15 | 上海百芝龙网络科技有限公司 | A kind of user's behavior prediction method in smart home |
| CN107180112A (en) * | 2017-06-15 | 2017-09-19 | 深圳市沃特沃德股份有限公司 | Using recommendation method and apparatus |
| CN107345860A (en) * | 2017-07-11 | 2017-11-14 | 南京康尼机电股份有限公司 | Rail vehicle door sub-health state recognition methods based on Time Series Data Mining |
| CN107679852A (en) * | 2017-09-28 | 2018-02-09 | 珠海市魅族科技有限公司 | Pay control method and device, terminal and readable storage medium storing program for executing |
| CN107864298A (en) * | 2017-12-25 | 2018-03-30 | 维沃移动通信有限公司 | A kind of intelligent prompt method and device |
| US10003924B2 (en) | 2016-08-10 | 2018-06-19 | Yandex Europe Ag | Method of and server for processing wireless device sensor data to generate an entity vector associated with a physical location |
| CN108920476A (en) * | 2018-03-30 | 2018-11-30 | 斑马网络技术有限公司 | Map retrieval calculates pass and holds up test macro and its test method |
| CN109005515A (en) * | 2018-09-05 | 2018-12-14 | 武汉大学 | A method of the user behavior pattern portrait based on motion track information |
| CN109146348A (en) * | 2017-06-27 | 2019-01-04 | 阿里巴巴集团控股有限公司 | A kind of logistics data processing method and processing device |
| CN109889672A (en) * | 2019-04-01 | 2019-06-14 | 中国科学技术大学 | Friend recommendation method based on mobile phone sensor |
| CN111695426A (en) * | 2020-05-08 | 2020-09-22 | 北京邮电大学 | Behavior pattern analysis method and system based on Internet of things |
| CN111770452A (en) * | 2020-05-27 | 2020-10-13 | 中山大学 | Mobile phone signaling stop point identification method based on personal travel track characteristics |
| CN112739984A (en) * | 2018-09-13 | 2021-04-30 | 华为技术有限公司 | Mobile phone multimodal position sensing |
| CN113283669A (en) * | 2021-06-18 | 2021-08-20 | 南京大学 | Intelligent planning travel investigation method and system combining initiative and passive |
| US11468536B2 (en) | 2018-05-18 | 2022-10-11 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for recommending a personalized pick-up location |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009043057A (en) * | 2007-08-09 | 2009-02-26 | Nomura Research Institute Ltd | Action history analysis device and method |
| CN201663612U (en) * | 2010-03-05 | 2010-12-01 | 东莞市华业龙图信息技术有限公司 | User interest modeling system based on location service |
| CN102036163A (en) * | 2009-10-02 | 2011-04-27 | 索尼公司 | Behaviour pattern analysis system, mobile terminal, behaviour pattern analysis method, and program |
| CN102682041A (en) * | 2011-03-18 | 2012-09-19 | 日电（中国）有限公司 | User behavior identification equipment and method |
-
2013
- 2013-04-22 CN CN2013101417262A patent/CN103218442A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009043057A (en) * | 2007-08-09 | 2009-02-26 | Nomura Research Institute Ltd | Action history analysis device and method |
| CN102036163A (en) * | 2009-10-02 | 2011-04-27 | 索尼公司 | Behaviour pattern analysis system, mobile terminal, behaviour pattern analysis method, and program |
| CN201663612U (en) * | 2010-03-05 | 2010-12-01 | 东莞市华业龙图信息技术有限公司 | User interest modeling system based on location service |
| CN102682041A (en) * | 2011-03-18 | 2012-09-19 | 日电（中国）有限公司 | User behavior identification equipment and method |
Cited By (59)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI589144B (en) * | 2013-10-21 | 2017-06-21 | Inventec Appliances Corp | Automatic cell phone context mode switching methods, controls and cell phones |
| CN103533171A (en) * | 2013-10-21 | 2014-01-22 | 英华达（上海）科技有限公司 | Method and control device for automatic mobile phone profile switching and mobile phone with automatic mobile phone profile switching function |
| CN103543832A (en) * | 2013-10-29 | 2014-01-29 | Tcl集团股份有限公司 | User identification method and device based on unbalanced data |
| CN103543832B (en) * | 2013-10-29 | 2017-08-04 | Tcl集团股份有限公司 | A kind of user identification method and device based on unbalanced data |
| US10423728B2 (en) | 2013-11-07 | 2019-09-24 | Huawei Technologies Co., Ltd. | Clustering method for a point of interest and related apparatus |
| CN104636354B (en) * | 2013-11-07 | 2018-02-06 | 华为技术有限公司 | A kind of position interest points clustering method and relevant apparatus |
| WO2015067119A1 (en) * | 2013-11-07 | 2015-05-14 | 华为技术有限公司 | Method for clustering position points of interest and related device |
| CN104636354A (en) * | 2013-11-07 | 2015-05-20 | 华为技术有限公司 | Position point of interest clustering method and related device |
| WO2015081480A1 (en) * | 2013-12-03 | 2015-06-11 | 深圳绿拓科技有限公司 | Room occupancy state sensing method and room occupancy state sensing apparatus |
| CN104731795A (en) * | 2013-12-19 | 2015-06-24 | 日本电气株式会社 | Mining apparatus and method of activity patterns of individuals |
| CN104731795B (en) * | 2013-12-19 | 2018-05-22 | 日本电气株式会社 | For excavating the device and method of individual activity pattern |
| CN103714139A (en) * | 2013-12-20 | 2014-04-09 | 华南理工大学 | Parallel data mining method for identifying a mass of mobile client bases |
| CN103714139B (en) * | 2013-12-20 | 2017-02-08 | 华南理工大学 | Parallel data mining method for identifying a mass of mobile client bases |
| CN105045792A (en) * | 2014-04-30 | 2015-11-11 | 三星电子株式会社 | Apparatus and method for integrated management of data in mobile device, and mobile device |
| US11403322B2 (en) | 2014-04-30 | 2022-08-02 | Samsung Electronics Co., Ltd. | Apparatus and method for integrated management of data in mobile device, and mobile device |
| CN104077404A (en) * | 2014-07-07 | 2014-10-01 | 西安交通大学 | Online taxpayer identity recognition method based on variable-length system calling sequence birthmarks |
| CN104077404B (en) * | 2014-07-07 | 2015-10-21 | 西安交通大学 | Based on people's identity ONLINE RECOGNITION method of declaring dutiable goods of elongated system call sequence birthmark |
| CN105388820A (en) * | 2014-08-25 | 2016-03-09 | 深迪半导体(上海)有限公司 | Intelligent monitoring device and monitoring method thereof, and monitoring system |
| CN104252527B (en) * | 2014-09-02 | 2018-04-20 | 百度在线网络技术（北京）有限公司 | A kind of method and apparatus of the resident information of definite mobile subscriber |
| CN104252527A (en) * | 2014-09-02 | 2014-12-31 | 百度在线网络技术（北京）有限公司 | Method and device for determining resident point information of mobile subscriber |
| CN105302900A (en) * | 2014-10-24 | 2016-02-03 | 康复广告株式会社 | Computer implemented method and system for social network service |
| CN105302900B (en) * | 2014-10-24 | 2018-07-10 | 康复广告株式会社 | For the method implemented by computer and system of social networking service |
| CN104615881A (en) * | 2015-01-30 | 2015-05-13 | 南京烽火星空通信发展有限公司 | User normal track analysis method based on movable position application |
| CN104615881B (en) * | 2015-01-30 | 2017-12-22 | 南京烽火星空通信发展有限公司 | A kind of user's normality trajectory analysis method based on shift position application |
| CN106022810A (en) * | 2015-02-13 | 2016-10-12 | 成都易驾科技有限公司 | Advertisement audience targeting and putting system based on driving behavior of car owner |
| CN106294485A (en) * | 2015-06-05 | 2017-01-04 | 华为技术有限公司 | Determine the method and device in notable place |
| CN106294485B (en) * | 2015-06-05 | 2019-11-01 | 华为技术有限公司 | Determine the method and device in significant place |
| DE102015212703B3 (en) * | 2015-07-07 | 2016-07-28 | Technische Universität Dresden | Method and apparatus for determining properties of at least one sub-micron structure |
| CN105307121B (en) * | 2015-10-16 | 2019-03-26 | 上海晶赞科技发展有限公司 | A kind of information processing method and device |
| CN105307121A (en) * | 2015-10-16 | 2016-02-03 | 上海晶赞科技发展有限公司 | Information processing method and device |
| CN105760837B (en) * | 2016-02-18 | 2019-01-04 | 西北工业大学 | Group mobility mode identification method based on heterogeneous sensor |
| CN105760837A (en) * | 2016-02-18 | 2016-07-13 | 西北工业大学 | Method for identifying group moving pattern based on heterogeneous sensor |
| US10003924B2 (en) | 2016-08-10 | 2018-06-19 | Yandex Europe Ag | Method of and server for processing wireless device sensor data to generate an entity vector associated with a physical location |
| CN106407277B (en) * | 2016-08-26 | 2019-10-25 | 北京车网互联科技有限公司 | It is a kind of based on car networking data to car owner's dwell point cluster after property analysis method |
| CN106407277A (en) * | 2016-08-26 | 2017-02-15 | 北京车网互联科技有限公司 | Internet of vehicles data-based attribute analysis method for vehicle owner parking point after being clustered |
| CN106339456A (en) * | 2016-08-26 | 2017-01-18 | 重庆科创职业学院 | Push method based on data mining |
| CN106384120A (en) * | 2016-08-29 | 2017-02-08 | 深圳先进技术研究院 | Mobile phone positioning data based resident activity pattern mining method and device |
| CN106384120B (en) * | 2016-08-29 | 2019-08-23 | 深圳先进技术研究院 | A kind of resident's activity pattern method for digging and device based on mobile phone location data |
| CN106708034A (en) * | 2016-11-23 | 2017-05-24 | 河池学院 | Implementation method for controlling walking path of robot on basis of mobile phone |
| CN107180112A (en) * | 2017-06-15 | 2017-09-19 | 深圳市沃特沃德股份有限公司 | Using recommendation method and apparatus |
| WO2018227760A1 (en) * | 2017-06-15 | 2018-12-20 | 深圳市沃特沃德股份有限公司 | Application recommendation method and device, and mobile terminal |
| CN109146348A (en) * | 2017-06-27 | 2019-01-04 | 阿里巴巴集团控股有限公司 | A kind of logistics data processing method and processing device |
| CN107345860A (en) * | 2017-07-11 | 2017-11-14 | 南京康尼机电股份有限公司 | Rail vehicle door sub-health state recognition methods based on Time Series Data Mining |
| CN107171872A (en) * | 2017-07-19 | 2017-09-15 | 上海百芝龙网络科技有限公司 | A kind of user's behavior prediction method in smart home |
| CN107679852A (en) * | 2017-09-28 | 2018-02-09 | 珠海市魅族科技有限公司 | Pay control method and device, terminal and readable storage medium storing program for executing |
| CN107864298A (en) * | 2017-12-25 | 2018-03-30 | 维沃移动通信有限公司 | A kind of intelligent prompt method and device |
| CN108920476B (en) * | 2018-03-30 | 2022-03-08 | 斑马网络技术有限公司 | Map retrieval route calculation engine test system and test method thereof |
| CN108920476A (en) * | 2018-03-30 | 2018-11-30 | 斑马网络技术有限公司 | Map retrieval calculates pass and holds up test macro and its test method |
| US11468536B2 (en) | 2018-05-18 | 2022-10-11 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for recommending a personalized pick-up location |
| CN109005515B (en) * | 2018-09-05 | 2020-07-24 | 武汉大学 | User behavior mode portrait drawing method based on movement track information |
| CN109005515A (en) * | 2018-09-05 | 2018-12-14 | 武汉大学 | A method of the user behavior pattern portrait based on motion track information |
| CN112739984A (en) * | 2018-09-13 | 2021-04-30 | 华为技术有限公司 | Mobile phone multimodal position sensing |
| CN109889672A (en) * | 2019-04-01 | 2019-06-14 | 中国科学技术大学 | Friend recommendation method based on mobile phone sensor |
| CN111695426A (en) * | 2020-05-08 | 2020-09-22 | 北京邮电大学 | Behavior pattern analysis method and system based on Internet of things |
| CN111695426B (en) * | 2020-05-08 | 2024-01-05 | 北京邮电大学 | Behavior pattern analysis method and system based on Internet of things |
| CN111770452A (en) * | 2020-05-27 | 2020-10-13 | 中山大学 | Mobile phone signaling stop point identification method based on personal travel track characteristics |
| CN111770452B (en) * | 2020-05-27 | 2021-06-01 | 中山大学 | Mobile phone signaling stop point identification method based on personal travel track characteristics |
| CN113283669A (en) * | 2021-06-18 | 2021-08-20 | 南京大学 | Intelligent planning travel investigation method and system combining initiative and passive |
| CN113283669B (en) * | 2021-06-18 | 2023-09-19 | 南京大学 | Active and passive combined intelligent planning travel investigation method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103218442A (en) | Method and system for life mode analysis based on mobile device sensor data | |
| Wang et al. | Applying mobile phone data to travel behaviour research: A literature review | |
| Toch et al. | Analyzing large-scale human mobility data: a survey of machine learning methods and applications | |
| CN108875007B (en) | method and device for determining interest point, storage medium and electronic device | |
| Xu et al. | A survey for mobility big data analytics for geolocation prediction | |
| Zhong et al. | Inferring building functions from a probabilistic model using public transportation data | |
| Orellana et al. | Exploring visitor movement patterns in natural recreational areas | |
| Frias-Martinez et al. | Characterizing urban landscapes using geolocated tweets | |
| Su et al. | Online travel mode identification using smartphones with battery saving considerations | |
| Montoliu et al. | Discovering places of interest in everyday life from smartphone data | |
| US8718672B2 (en) | Identifying status based on heterogeneous sensors | |
| Kounadi et al. | Population at risk: Using areal interpolation and Twitter messages to create population models for burglaries and robberies | |
| Abdulazim et al. | Using smartphones and sensor technologies to automate collection of travel data | |
| Ferrari et al. | Discovering daily routines from google latitude with topic models | |
| CN104737523A (en) | Managing a context model in a mobile device by assigning context labels for data clusters | |
| CN105307121B (en) | A kind of information processing method and device | |
| Espín Noboa et al. | Discovering and characterizing mobility patterns in urban spaces: A study of manhattan taxi data | |
| Higuchi et al. | Mobile devices as an infrastructure: A survey of opportunistic sensing technology | |
| CN106339456A (en) | Push method based on data mining | |
| Umair et al. | Discovering personal places from location traces | |
| Bwambale et al. | Getting the best of both worlds: a framework for combining disaggregate travel survey data and aggregate mobile phone data for trip generation modelling | |
| Hang et al. | Platys: User-centric place recognition | |
| Zhang et al. | Adaptive learning of semantic locations and routes | |
| Elnour et al. | Social Internet of Things (SIoT) Localization for Smart Cities Traffic Applications | |
| Wang | Understanding activity location choice with mobile phone data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20130724 |