CN107563402A - A kind of social networks estimating method and system - Google Patents
A kind of social networks estimating method and system Download PDFInfo
- Publication number
- CN107563402A CN107563402A CN201710552281.5A CN201710552281A CN107563402A CN 107563402 A CN107563402 A CN 107563402A CN 201710552281 A CN201710552281 A CN 201710552281A CN 107563402 A CN107563402 A CN 107563402A
- Authority
- CN
- China
- Prior art keywords
- user
- mobile
- space
- intention
- occurrence
- 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
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention, which provides a kind of social networks estimating method and system, methods described, to be included：Based on tensor resolution method, obtained from space-time data in the same period, multiple mobile intentions of the multiple mobile intentions and second user of the first user；Based on default multi classifier, from multiple mobile intentions of the multiple mobile intentions and second user of first user, the mobile intention of first user and the second user co-occurrence is obtained；Mobile intention based on the disaggregated models of SVM bis- and the co-occurrence, infer the social networks of first user and the second user.Social networks estimating method provided by the invention and system, being decomposited by tensor resolution method has the mobile intention of same space-time characterisation in user's space-time data, so as to establish the mapping relations between space-time data and the mobile intention, and then infer the social relationships between user, improve the accuracy rate of deduction.
Description
Technical field
The present invention relates to areas of information technology, more particularly, to a kind of social networks estimating method and system.
Background technology
At present, as the Intelligent mobile equipment with multiple sensors such as mobile phone, tablet personal computer is widely available, social networks
Fast development, increasing space-time data can be collected into, such as：Person-to-person communication data, GPS track, diverse location clothes
Data of registering that business business provides etc..And with the development of Internet of Things, being originally used for the system of other purposes also further increases
The capacity gauge of space-time data, such as Intelligent bus card system, video monitoring system, ATM system etc..From these space-time datas
Social relationships between middle excavation people and people are most important for some popular applications, such as friend recommendation, advertisement putting, stream
The sprawling of row disease, crime or the determination of terroristic organization member etc..
In order to be inferred to the social relationships of two people from space-time data, prior art is pushed away by the co-occurrence feature of two people
The tightness degree of disconnected two personal relationships.The co-occurrence of two people refers to that two people appear in the phenomenon in identical place in the same time.
In general, more close two people of relation, the number of co-occurrence are more, then directly can be judged by the number number of co-occurrence
The close relation degree of two people.The place of co-occurrence can determine public place or private site according to the entropy in place,
The calculation formula of place entropy is as follows：
Wherein, Pu,lWhat is represented is that some user appears in the number in l-th of place and accounts for all users and appear at l-th
The ratio of place total degree.If the entropy in place is bigger, then represents that many people arrived the place, then the place is a public affairs
Place altogether.So co-occurrence in the higher place of entropy is less to the relation contribution between two people, and the co-occurrence in the small place of entropy is to two
The relation contribution of people is larger, then may infer that the number of co-occurrence occur by entropy, so as to infer the relation of two people.
But the people of two close relations also can in the high local co-occurrence of entropy, such as the position such as megastore, supermarket, and
Most data sources in existing space-time data are in public place, therefore the space-time data in the small place of entropy is not easy to obtain
Take, therefore in actual applications, it is existing to infer that the method for social networks is inaccurate according to entropy.
The content of the invention
The present invention provides a kind of a kind of social networks for overcoming above mentioned problem or solving the above problems at least in part and pushed away
Disconnected method and system.
According to provided by the invention in a first aspect, of the invention provide a kind of social networks estimating method, methods described includes：
S1, based on tensor resolution method, obtained from space-time data in the same period, the multiple of the first user mobile are intended to
And multiple mobile intentions of second user；
S2, based on default multi classifier, from first user it is multiple it is mobile be intended to and second user it is more
In individual mobile intention, the mobile intention of first user and the second user co-occurrence is obtained；
S3, the mobile intention based on the disaggregated models of SVM bis- and the co-occurrence, infer that first user and described second uses
The social networks at family.
Wherein, step S1 includes：
By the space-time data, according to the default period, multiple space-time subdatas are divided into；
Calculate in each period, the tensor element value of the space-time subdata；
Based on the characteristic vector of the tensor element value, the multiple mobile intentions and second for extracting first user are used
Multiple mobile intentions at family.
Wherein, step S1 foregoing description method also includes：
The space-time data is obtained, and using the partial data of the space-time data as training dataset, to the multiclass
Grader is trained.
Wherein, it is described to obtain the space-time data, and using the partial data of the space-time data as training dataset, it is right
The multi classifier is trained, including：
Based on the tensor element value, the characteristic of division of the multi classifier is determined；
Based on the characteristic of division, using the training dataset, the multi classifier is trained.
Wherein, step S2 includes：
Multiple mobile characteristic of division values for being intended to, calculating first user based on first user；
Multiple mobile characteristic of division values for being intended to, calculating the second user based on the second user；
Based on the multi classifier, the classification for comparing the characteristic of division value and the second user of first user is special
Value indicative, extract the mobile intention of first user and the second user co-occurrence.
Wherein, the characteristic of division includes：
The one or more of space characteristics, temporal characteristics and date feature.
Wherein, step S3 includes：
By first user and the mobile intention of the second user co-occurrence, multidimensional characteristic vectors are converted into；
Based on the disaggregated models of SVM bis- and the multidimensional characteristic vectors, first user and second user are inferred
Social relationships.
According to the second aspect of the invention, there is provided a kind of social networks inference system, including：
Tensor resolution module, for based on tensor resolution method, being obtained from space-time data in the same period, the first user
Multiple mobile be intended to and the multiple of second user mobile are intended to；
Multi classifier module, for based on default multi classifier, being intended to from multiple movements of first user
And in multiple mobile intentions of second user, obtain the mobile intention of first user and the second user co-occurrence；
Inference module, for the mobile intention based on the disaggregated models of SVM bis- and the co-occurrence, infer first user and
The social networks of the second user.
According to the third aspect of the invention we, there is provided a kind of computer program product, including program code, described program code
For performing social networks estimating method described above.
According to the fourth aspect of the invention, there is provided a kind of non-transient computer readable storage medium storing program for executing, for storing such as preceding institute
The computer program stated.
Social networks estimating method provided by the invention and system, are decomposited in user's space-time data by tensor resolution method
Mobile intention with same space-time characterisation, so as to establish the mapping relations between space-time data and the mobile intention, and then
Infer the social relationships between user, improve the accuracy rate of deduction.
Brief description of the drawings
Fig. 1 is a kind of social networks estimating method flow chart provided in an embodiment of the present invention；
Fig. 2 is social networks estimating method comparison of test results figure provided in an embodiment of the present invention；
Fig. 3 is a kind of social networks inference system structure chart provided in an embodiment of the present invention.
Embodiment
With reference to the accompanying drawings and examples, the embodiment of the present invention is described in further detail.Implement below
Example is used to illustrate the present invention, but is not limited to the scope of the present invention.
Fig. 1 is a kind of social networks estimating method flow chart provided in an embodiment of the present invention, as shown in figure 1, including：
S1, based on tensor resolution method, obtained from space-time data in the same period, the multiple of the first user mobile are intended to
And multiple mobile intentions of second user；
S2, based on default multi classifier, from first user it is multiple it is mobile be intended to and second user it is more
In individual mobile intention, the mobile intention of first user and the second user co-occurrence is obtained；
S3, the mobile intention based on the disaggregated models of SVM bis- and the co-occurrence, infer that first user and described second uses
The social networks at family.
In the S1, the tensor resolution method is CP decomposition methods, i.e., by a tensor representation into the limited individual tensor of order one it
With the CP decomposition methods can be decomposed effectively to the space-time characterisation in space-time data, based on the tensor resolution method, energy
Enough Move Modes that user is extracted from space-time data, i.e. user different time sections mobile intention, it is described it is mobile be intended to be
For the mobile purpose.
In the S2, it is to be understood that the multi classifier is to be used to establish between user record and mobile purpose
Mapping function, for it a variety of mobile purposes be present in the case of, can will see a multiclass as to the classification of mobile purpose
Classification, so set up the mobile purpose feature required for multi classifier determines.
In the S3, the disaggregated models of SVM bis- are traditional classifier, are very effective linear classifiers, according to altogether
Existing destination information, it can infer that the first user and second user belong to stranger or acquaintance's class by linear classifier
Not.
Social networks estimating method provided by the invention and system, are decomposited in user's space-time data by tensor resolution method
Mobile destination with same space-time characterisation, so as to establish the mapping relations between space-time data and the mobile destination,
And then infer the social relationships between user, improve the accuracy rate of reckoning.
On the basis of above-described embodiment, step S1 includes：
By the space-time data, according to the default period, multiple space-time subdatas are divided into；
Calculate in each period, the tensor element value of the space-time subdata；
Based on the characteristic vector of the tensor element value, the multiple mobile intentions and second for extracting first user are used
Multiple mobile intentions at family.
Specifically, for example：Data are obtained using the data of Beijing Public Transport all-purpose card, the Data Collection is 2014
The record of swiping the card of the main bus in areas of Beijing on October 31 in 1 day to 2014 October.
Using tensor resolution algorithm, the mobile purpose that data are concentrated is extracted.Calculate each tensor elementValue, meter
Calculation method is as follows：
All unique bus stations are used as place, a corresponding r using in datai.Do not surpass if Liang Ge bus stations are separated by
300 meters are crossed, then is considered as a bus station.It was divided into following eight periods by one day, the time of each period is as follows：
00:00:00-4:59:59、05:00:00-6:59:59、07:00:00-8:59:59、09:00:00-10:59:59、
11:00:00-13:59:59、14:00:00-16:59:59、17:00:00-19:59:59、20:00:00-23:59:59。
Count all riderships of on October 31 October 1 daily each period aboveAnd
Each bus station riRidership Count (ri,tj,dk), then each tensor elementValue be：
After the value of all tensor elements is obtained, using CP decomposing programs, tensor Y is subjected to CP decomposition：
Wherein, λrIt is coefficient, YrIt is obtained single order tensor, can be write as three vectorial ar,br,crApposition.
Wherein, ar,br,crIt is place respectively, the characteristic vector on time and date；And YrIt is that we take out from data set
The mobile purpose taken.On the data set for Beijing's Bus Card that we use, we have extracted 7 kinds of mobile mesh
's.
It should be noted that areas of Beijing on October 31,1 day to 2014 October in 2014 provided in an embodiment of the present invention
7 kinds of mobile purposes of the record data extraction of swiping the card of main bus be only have in the program inherently move purpose, this hair
Bright embodiment concentrates the mobile purpose quantity possessed to be not specifically limited different pieces of information.
The user that the embodiment of the present invention extracts space-time data concentration by using the method for tensor resolution moves purpose, as
Differentiate feature, obtain conveniently and feature is obvious, the social relationships between subsequent user, which are inferred, provides differentiation basis.
On the basis of above-described embodiment, step S1 foregoing description methods also include：
The space-time data is obtained, and using the partial data of the space-time data as training dataset, to the multiclass
Grader is trained.
Specifically, the space-time data that we are obtained is concentrated, according to the when and where of record, we identify 412
Card user, and by inquiry, it is determined that the users of 2796 pairs of understanding, while using existing algorithm extract 412 users it
Between all co-occurrence, the co-occurrence includes understanding and do not recognize all co-occurrences between user.We recognize these determinations
2796 couples of users in pick out 80% user at random to as training data, a multi classifier being trained, to establish use
Family records the mapping function between mobile purpose.
It is described to obtain the space-time data on the basis of above-described embodiment, and by the partial data of the space-time data
Collection is used as training dataset, and the multi classifier is trained, including：
Based on the tensor element value, the characteristic of division of the multi classifier is determined；
Based on the characteristic of division, using the training dataset, the multi classifier is trained.
Based on the tensor element value of tensor resolution during S1, tensor information can be divided into multiclass feature, then each
Any mobile purpose corresponding to record, it is mapped in equivalent to corresponding in the mobile purpose of which kind, further according to feature pair
It is classified.
On the basis of above-described embodiment, step S2 includes：
Multiple mobile characteristic of division values for being intended to, calculating first user based on first user；
Multiple mobile characteristic of division values for being intended to, calculating the second user based on the second user；
Based on the multi classifier, the classification for comparing the characteristic of division value and the second user of first user is special
Value indicative, extract the mobile intention of first user and the second user co-occurrence.
Specifically, the embodiment of the present invention uses existing Adaboost algorithm and training dataset, Adaboost is trained
Model.According to the Adaboost disaggregated models trained, the destination information of these co-occurrences is mapped to mobile purpose pair.This hair
Seven kinds of mobile purposes are extracted in bright embodiment, 7 kinds of different mobile purposes pair can be produced, the mobile purpose is to being to extract
Co-occurrence feature.
On the basis of above-described embodiment, the characteristic of division includes：
The one or more of space characteristics, temporal characteristics and date feature.
The information that single order tensor according to decompositing before provides, and we be made some Feature Engineerings experiment, we
Three major types feature is determined, is space characteristics, temporal characteristics and date feature respectively, classification provided in an embodiment of the present invention is special
Sign may include the one or more of this three category feature, and this three category feature specifically includes：
Space characteristics：Wherein, the space characteristics include place entropy, and the place entropy is used for weighing the temperature in place,
Its calculation formula is as follows：
Wherein piIt is that i-th of user appears in lockNumber account for all users and appear in lockNumber ratio.
The space characteristics further comprises venue type, and the venue type represents the classification belonging to place, such as bar,
Supermarket etc..Venue type can be obtained by some application programming interfaces based on location-based service.
The space characteristics further comprises distance, and the distance is represented on distance location in current record in a record
Distance location.
Temporal characteristics：Wherein, the temporal characteristics include hour, and the hour represents the time of record, value from 0 to
23。
The temporal characteristics further comprises the pseudo- residence time, the pseudo- residence time represent time in current record with it is upper
Time difference in one record.
The temporal characteristics further comprises time interval, and the time interval is used for weighing some place, some user
Access habits.Equal to the time in current record and time difference in the record in a upper identical place.
The temporal characteristics also include secondary time interval, and the secondary time interval is equally used for weighing some ground
Point, the access habits of some user.Equal to the time in current record and time difference in preceding second record in identical place.
Date feature：Wherein, the date feature includes week, represents Sunday, Monday to week with 0 to 6 in the week
Six.
The date feature also includes the date, the number of days in month where the date is represented with 0 to 31.
The date feature also includes date type, and the date type represents the classification on date, and the embodiment of the present invention carries
Three kinds of date categories are supplied：At weekend, small long holidays and long holidays, corresponding respectively is vacation two day weekend, the Mid-autumn Festival this three to five
It vacation, and more than five days and the vacation of the above.
The embodiment of the present invention is by the way that characteristic of division specifically to be divided into space characteristics, temporal characteristics and date feature, to dividing
Class is refined so that classification is more accurate.
On the basis of above-described embodiment, step S3 includes：
By first user and the mobile intention of the second user co-occurrence, multidimensional characteristic vectors are converted into；
Based on the disaggregated models of SVM bis- and the multidimensional characteristic vectors, first user and second user are inferred
Social relationships.
Specifically, for example：The movement of the first user and the second user co-occurrence that obtain are intended to above-described embodiment and carried
The data of confession, the Information Number of resolving is 7 kinds, then can produceThe different mobile purpose pair of kind.Count two users
All mobile purposes pair, form the characteristic vectors of 21 dimensions, the disaggregated models of SVM bis- trained using 21 dimensional feature vector,
Judge that first user and the second user are stranger or acquaintance further according to 21 dimensional feature vectors.
Fig. 2 is social networks estimating method inferred results comparison diagram provided in an embodiment of the present invention, as shown in Fig. 2 black
Solid line is the inferred results of the embodiment of the present invention, and black dotted lines are the inferred results of prior art, figure it is seen that pushing away
In disconnected accuracy rate, the method based on place entropy co-occurrence that the embodiment of the present invention will be substantially better than prior art is inferred.
Social networks estimating method provided by the invention and system, are decomposited in user's space-time data by tensor resolution method
Mobile destination with same space-time characterisation, so as to establish the mapping relations between space-time data and the mobile destination,
And then infer the social relationships between user, improve the accuracy rate of reckoning.
Fig. 3 is a kind of social networks inference system structure chart provided in an embodiment of the present invention, including：Tensor resolution module 1,
Multi classifier module 2 and inference module 3, wherein,
Tensor resolution module 1 is used to be based on tensor resolution method, is obtained from space-time data in the same period, the first user
Multiple mobile be intended to and the multiple of second user mobile are intended to；
Multi classifier module 2 is used to be based on default multi classifier, from multiple mobile intentions of first user
And in multiple mobile intentions of second user, obtain the mobile intention of first user and the second user co-occurrence；
Inference module 3 is used for the mobile intention based on the disaggregated models of SVM bis- and the co-occurrence, infer first user and
The social networks of the second user.
Specific social networks estimating method can be found in above-described embodiment, and the embodiment of the present invention will not be repeated here.
The present embodiment provides a kind of social networks inference system, including：At least one processor；And with the processor
At least one memory of communication connection, wherein：
The memory storage has and by the programmed instruction of the computing device, the processor described program can be called to refer to
Order to perform the method that above-mentioned each method embodiment is provided, such as including：S1, based on tensor resolution method, from space-time data
Obtain in the same period, multiple mobile intentions of the multiple mobile intentions and second user of the first user；S2, based on default
Multi classifier, from first user it is multiple it is mobile be intended to and second user it is multiple it is mobile be intended to, obtain institute
State the mobile intention of the first user and the second user co-occurrence；S3, the mobile meaning based on the disaggregated models of SVM bis- and the co-occurrence
Figure, infer the social networks of first user and the second user.
The present embodiment discloses a kind of computer program product, and the computer program product includes being stored in non-transient calculating
Computer program on machine readable storage medium storing program for executing, the computer program include programmed instruction, when described program instruction is calculated
When machine performs, computer is able to carry out the method that above-mentioned each method embodiment is provided, such as including：S1, based on tensor resolution
Method, obtained from space-time data in the same period, multiple movements of the multiple mobile intentions and second user of the first user
It is intended to；S2, based on default multi classifier, multiple from first user mobile are intended to and multiple shiftings of second user
In dynamic intention, the mobile intention of first user and the second user co-occurrence is obtained；S3, based on the disaggregated models of SVM bis- and
The mobile intention of the co-occurrence, infer the social networks of first user and the second user.
The present embodiment provides a kind of non-transient computer readable storage medium storing program for executing, the non-transient computer readable storage medium storing program for executing
Computer instruction is stored, the computer instruction makes the computer perform the method that above-mentioned each method embodiment is provided, example
Such as include：S1, based on tensor resolution method, obtained from space-time data in the same period, the multiple of the first user mobile are intended to
And multiple mobile intentions of second user；S2, based on default multi classifier, from multiple mobile meanings of first user
In multiple mobile intentions of figure and second user, the mobile intention of first user and the second user co-occurrence is obtained；
S3, the mobile intention based on the disaggregated models of SVM bis- and the co-occurrence, infer the social activity of first user and the second user
Relation.
One of ordinary skill in the art will appreciate that：Realizing all or part of step of above method embodiment can pass through
Programmed instruction related hardware is completed, and foregoing program can be stored in a computer read/write memory medium, the program
Upon execution, the step of execution includes above method embodiment；And foregoing storage medium includes：ROM, RAM, magnetic disc or light
Disk etc. is various can be with the medium of store program codes.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
Realized by the mode of software plus required general hardware platform, naturally it is also possible to pass through hardware.Based on such understanding, on
The part that technical scheme substantially in other words contributes to prior art is stated to embody in the form of software product, should
Computer software product can store in a computer-readable storage medium, such as ROM/RAM, magnetic disc, CD, including some fingers
Make to cause a computer equipment (can be personal computer, server, or network equipment etc.) to perform each implementation
Method described in some parts of example or embodiment.
Finally, the present processes are only preferable embodiment, are not intended to limit the scope of the present invention.It is all
Within the spirit and principles in the present invention, any modification, equivalent substitution and improvements made etc., the protection of the present invention should be included in
Within the scope of.
Claims (10)
- A kind of 1. social networks estimating method, it is characterised in that including：S1, based on tensor resolution method, obtained from space-time data in the same period, the first user it is multiple it is mobile be intended to and Multiple mobile intentions of second user；S2, based on default multi classifier, multiple from first user mobile are intended to and multiple shiftings of second user In dynamic intention, the mobile intention of first user and the second user co-occurrence is obtained；S3, the mobile intention based on the disaggregated models of SVM bis- and the co-occurrence, deduction first user and the second user Social networks.
- 2. according to the method for claim 1, it is characterised in that step S1 includes：By the space-time data, according to the default period, multiple space-time subdatas are divided into；Calculate in each period, the tensor element value of the space-time subdata；Based on the characteristic vector of the tensor element value, the multiple mobile intentions and second user of first user are extracted Multiple mobile intentions.
- 3. according to the method for claim 2, it is characterised in that step S1 foregoing description methods also include：The space-time data is obtained, and using the partial data of the space-time data as training dataset, to the multicategory classification Device is trained.
- 4. according to the method for claim 3, it is characterised in that it is described to obtain the space-time data, and by the space-time number According to partial data as training dataset, the multi classifier is trained, including：Based on the tensor element value, the characteristic of division of the multi classifier is determined；Based on the characteristic of division, using the training dataset, the multi classifier is trained.
- 5. according to the method for claim 4, it is characterised in that step S2 includes：Multiple mobile characteristic of division values for being intended to, calculating first user based on first user；Multiple mobile characteristic of division values for being intended to, calculating the second user based on the second user；Based on the multi classifier, the characteristic of division value of first user and the characteristic of division of the second user are compared Value, extract the mobile intention of first user and the second user co-occurrence.
- 6. according to the method for claim 5, it is characterised in that the characteristic of division includes：The one or more of space characteristics, temporal characteristics and date feature.
- 7. any method according to claim 1-6, it is characterised in that step S3 includes：By first user and the mobile intention of the second user co-occurrence, multidimensional characteristic vectors are converted into；Based on the disaggregated models of SVM bis- and the multidimensional characteristic vectors, the society of deduction first user and second user Relation.
- A kind of 8. social networks inference system, it is characterised in that including：Tensor resolution module, for based on tensor resolution method, being obtained from space-time data in the same period, the first user's is more Multiple mobile intentions of individual mobile intention and second user；Multi classifier module, for based on default multi classifier, from first user it is multiple it is mobile be intended to and In multiple mobile intentions of second user, the mobile intention of first user and the second user co-occurrence is obtained；Inference module, for the mobile intention based on the disaggregated models of SVM bis- and the co-occurrence, infer first user and described The social networks of second user.
- 9. a kind of computer program product, it is characterised in that the computer program product includes being stored in non-transient computer Computer program on readable storage medium storing program for executing, the computer program include programmed instruction, when described program is instructed by computer During execution, the computer is set to perform the method as described in claim 1 to 7 is any.
- 10. a kind of non-transient computer readable storage medium storing program for executing, it is characterised in that the non-transient computer readable storage medium storing program for executing is deposited Computer instruction is stored up, the computer instruction makes the computer perform the method as described in claim 1 to 7 is any.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710552281.5A CN107563402A (en) | 2017-07-07 | 2017-07-07 | A kind of social networks estimating method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710552281.5A CN107563402A (en) | 2017-07-07 | 2017-07-07 | A kind of social networks estimating method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107563402A true CN107563402A (en) | 2018-01-09 |
Family
ID=60973071
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710552281.5A Pending CN107563402A (en) | 2017-07-07 | 2017-07-07 | A kind of social networks estimating method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107563402A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110675192A (en) * | 2019-09-27 | 2020-01-10 | 深圳市掌众信息技术有限公司 | Intimacy mining method, advertisement pushing method and system |
| CN111553386A (en) * | 2020-04-07 | 2020-08-18 | 哈尔滨工程大学 | AdaBoost and CNN-based intrusion detection method |
| CN113115200A (en) * | 2019-12-24 | 2021-07-13 | 中国移动通信集团浙江有限公司 | User relationship identification method and device and computing equipment |
-
2017
- 2017-07-07 CN CN201710552281.5A patent/CN107563402A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110675192A (en) * | 2019-09-27 | 2020-01-10 | 深圳市掌众信息技术有限公司 | Intimacy mining method, advertisement pushing method and system |
| CN113115200A (en) * | 2019-12-24 | 2021-07-13 | 中国移动通信集团浙江有限公司 | User relationship identification method and device and computing equipment |
| CN113115200B (en) * | 2019-12-24 | 2023-04-18 | 中国移动通信集团浙江有限公司 | User relationship identification method and device and computing equipment |
| CN111553386A (en) * | 2020-04-07 | 2020-08-18 | 哈尔滨工程大学 | AdaBoost and CNN-based intrusion detection method |
| CN111553386B (en) * | 2020-04-07 | 2022-05-20 | 哈尔滨工程大学 | AdaBoost and CNN-based intrusion detection method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Pei et al. | A new insight into land use classification based on aggregated mobile phone data | |
| CN102708130B (en) | Calculate the easily extensible engine that fine point of user is mated for offer | |
| Allahviranloo et al. | Daily activity pattern recognition by using support vector machines with multiple classes | |
| Sewell et al. | Latent space models for dynamic networks with weighted edges | |
| CN107767021A (en) | A kind of risk control method and equipment | |
| US8255392B2 (en) | Real time data collection system and method | |
| WO2020038100A1 (en) | Feature relationship recommendation method and apparatus, computing device and storage medium | |
| CN110443120A (en) | A kind of face identification method and equipment | |
| CN111695019B (en) | Method and device for identifying associated account | |
| CN107563402A (en) | A kind of social networks estimating method and system | |
| CN108647800A (en) | A kind of online social network user missing attribute forecast method based on node insertion | |
| Zhuang et al. | From compound word to metropolitan station: Semantic similarity analysis using smart card data | |
| Chen et al. | Unraveling latent transfer patterns between metro and bus from large-scale smart card data | |
| CN113435900A (en) | Transaction risk determination method and device and server | |
| CN113761359A (en) | Data packet recommendation method and device, electronic equipment and storage medium | |
| CN114004654A (en) | Internet card user portrait construction method and system based on big data analysis | |
| CN115375205A (en) | Method, device and equipment for determining water user portrait | |
| Liao et al. | GRBMC: An effective crowdsourcing recommendation for workers groups | |
| Sun et al. | Deep convolutional autoencoder for urban land use classification using mobile device data | |
| Zhan et al. | A semantic sequential correlation based lstm model for next poi recommendation | |
| Shen et al. | Novel model for predicting individuals’ movements in dynamic regions of interest | |
| CN109902129A (en) | Insurance agent's classifying method and relevant device based on big data analysis | |
| CN111797856B (en) | Modeling method and device, storage medium and electronic equipment | |
| Wang et al. | A Comparative Study on Contract Recommendation Model: Using Macao Mobile Phone Datasets | |
| Xu et al. | Mining urban lifestyles: urban computing, human behavior and recommender systems |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180109 |
|
| RJ01 | Rejection of invention patent application after publication |