CN107316044A - Similar users recognition methods and device - Google Patents
Similar users recognition methods and device Download PDFInfo
- Publication number
- CN107316044A CN107316044A CN201610268871.0A CN201610268871A CN107316044A CN 107316044 A CN107316044 A CN 107316044A CN 201610268871 A CN201610268871 A CN 201610268871A CN 107316044 A CN107316044 A CN 107316044A
- Authority
- CN
- China
- Prior art keywords
- user
- base station
- index
- predetermined period
- similarity
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The present invention proposes a kind of similar users recognition methods and device, is related to data analysis field.Wherein, a kind of similar users recognition methods of the invention includes：Extract position and time that user's communication behavior occurs；According to user within a predetermined period of time communication behavior occur position and Time Calculation user relative to base station base station common index；Predetermined quantity base station is extracted relative to the base station common index of different base station according to user, generation user often uses base station characteristic vector；User's index similarity is often determined with base station characteristic vector according to the user of different user.Pass through such method, the position and time that can be occurred based on communication behavior are obtained user and often use base station characteristic vector, often user's index similarity is calculated further according to user with base station characteristic vector, with good versatility, improve the degree of accuracy of Similarity Measure, computational complexity is relatively low, improves operation efficiency, reduces the performance requirement to arithmetic facility.
Description
Technical field
The present invention relates to data analysis field, particularly a kind of similar users recognition methods and dress
Put.
Background technology
User's similarity analysis is one of user behavior analysis processing more popular aspect, mesh
Preceding industrial customers similarity identification technical sophistication is various, including passes through mobile phone IMEI
(International Mobile Equipment Identity, International Mobile Station Equipment Identification) is right
Relationship cycle cosine phase is realized than analysis identification, and based on user's calling behavior frequency dependence index
Calculated like degree.
But, the applicable surface based on mobile phone IMEI comparative analyses identification is small, accuracy rate is low；And
Relationship cycle cosine similarity based on calling behavior frequency is calculated according only to the detailed forms data of user's communication
Call frequency is calculated, the degree of accuracy is low, it is impossible to play effective recognition reaction, and cosine
Similarity Measure complexity is high, efficiency is low, is unfavorable for expanding application.A kind of high relevance grade, height
The degree of accuracy and efficient user's similarity analysis method are that the urgent of ownership's behavioural analysis is essential
Ask.
The content of the invention
It is an object of the present invention to improve the efficiency of user's similarity analysis, the degree of accuracy and logical
The property used.
According to an aspect of the present invention, a kind of similar users recognition methods is proposed, including：Carry
Take position and time that family communication behavior occurs；Communicated within a predetermined period of time row according to user
The base station common index of position and Time Calculation user relative to base station for generation；According to user
Base station common index relative to different base station extracts predetermined quantity base station, generation user Chang Yongji
Stand characteristic vector；User's similarity is often determined with base station characteristic vector according to the user of different user
Index.
Alternatively, user's communication behavior includes conversing and/or connection data network；User communicates
The base station that the position that behavior occurs is interacted when occurring communication behavior for user.
Alternatively, counted according to the user position that communication behavior occurs within a predetermined period of time and time
Calculate user includes relative to the base station common index of base station：Within a predetermined period of time, obtain predetermined
Cycle base station common index, wherein, predetermined period base station common index is that user occurs with base station
Cross the ratio of the number of predetermined period in the number and predetermined amount of time of interactive predetermined period；In advance
Fixed cycle includes one month, ten days, seven days, three days and/or one day；According to predetermined period base station
Common index determines base station common index.
Alternatively, obtaining predetermined period base station common index includes：According to formula
Predetermined period base station common index is determined, wherein, I is predetermined period base station common index,
N is the number of predetermined period in predetermined amount of time, and i is the predetermined period mark in predetermined amount of time
Number, tiWhether occurred the mark interacted with base station for user in i-th of predetermined period：If
User occurred to interact with base station in i predetermined period, then tiFor 1；If i-th of predetermined week
User did not occurred to interact with base station in phase, then tiFor 0.
Alternatively, determine that base station common index includes according to predetermined period base station common index：Root
According to formula
CI=35My+30Ty+16Wy+12THy+8Dy
Base station common index is determined, wherein, CI is that the base station of user within a predetermined period of time is normal
With index, MyFor the predetermined period base station common index that predetermined period is one month；TyIt is predetermined
Cycle is the predetermined period base station common index of ten days；WyFor the predetermined week that predetermined period is seven
Phase base station common index；THyFor the predetermined period base station common index that predetermined period is three；
DyFor the predetermined period base station common index that predetermined period is one.
Alternatively, user's similarity is often determined with base station characteristic vector according to the user of different user
Index includes：According to formula
S=(Pm∩Pn)/Y
Determine user's index similarity, wherein, m, n be user mark, S be user n with
User m index similarity, PmFor user m conventional base station characteristic vector, PnFor user
N conventional base station characteristic vector, Pm∩PnFor PmWith PnMiddle identical base station number, Y is
User often uses the quantity of base station in the characteristic vector of base station.
Alternatively, the position and time for extracting the generation of user's communication behavior include：Extract user's work
Make position and time that the communication behavior of day occurs；Communicated within a predetermined period of time row according to user
Position and Time Calculation user for generation include relative to the base station common index of base station：According to
The user position that workaday communication behavior occurs in predetermined amount of time and Time Calculation user's phase
For the working day base station common index of base station；It is normal relative to the base station of different base station according to user
With exponent extracting predetermined quantity base station, generation user is often included with base station characteristic vector：According to
Predetermined quantity base station is extracted in family relative to the size of the working day base station common index of different base station,
Generate the conventional base station characteristic vector of user job day；It is often special with base station according to the user of different user
Levy vector and determine that user's index similarity includes：According to the conventional base of the user job of different user day
Characteristic vector of standing determines user's index similarity.
Alternatively, the position and time for extracting the generation of user's communication behavior include：Extract user's section
Position and time that the communication behavior of holiday occurs；Communicated within a predetermined period of time row according to user
Position and Time Calculation user for generation include relative to the base station common index of base station：According to
Position and Time Calculation user's phase that the communication behavior of user's festivals or holidays in predetermined amount of time occurs
For the festivals or holidays base station common index of base station；It is normal relative to the base station of different base station according to user
With exponent extracting predetermined quantity base station, generation user is often included with base station characteristic vector：According to
Predetermined quantity base station is extracted in family relative to the size of the festivals or holidays base station common index of different base station,
Generation user's festivals or holidays often use base station characteristic vector；It is often special with base station according to the user of different user
Levy vector and determine that user's index similarity includes：Often base is used according to user's festivals or holidays of different user
Characteristic vector of standing determines user's index similarity.
Alternatively, counted according to the user position that communication behavior occurs within a predetermined period of time and time
Calculate user includes relative to the base station common index of base station：According to user in predetermined amount of time work
Make day communication behavior occur position and Time Calculation user relative to base station working day base station
Common index；According to the communication behavior of user's festivals or holidays in predetermined amount of time occur position and
Festivals or holidays base station common index of the Time Calculation user relative to base station；According to user relative to not
Base station common index with base station extracts predetermined quantity base station, generation user often with base station feature to
Amount includes：Extracted according to user relative to the size of the working day base station common index of different base station
Predetermined quantity base station, the conventional base station characteristic vector of generation user job day；According to user relative to
The size of the festivals or holidays base station common index of different base station extracts predetermined quantity base station, generates user
Festivals or holidays often use base station characteristic vector；It is often true with base station characteristic vector according to the user of different user
Determining user's index similarity includes：According to day conventional base station feature of the user job of different user to
Amount determines user job day index similarity；Often base station is used according to user's festivals or holidays of different user
Characteristic vector determines user's festivals or holidays index similarity；Based on user job day index similarity and
User's festivals or holidays index similarity determines user's comprehensive similarity index.
Alternatively, in addition to：Index similarity is compared with predetermined threshold；If similarity refers to
Number is not less than predetermined threshold, it is determined that user is similar users；If index similarity is less than predetermined
Threshold value, it is determined that user is non-similar users.
By such method, the position and time that can be occurred based on communication behavior obtain user
Conventional base station characteristic vector, often calculates user's similarity with base station characteristic vector further according to user and refers to
Number, with good versatility；Due to consideration that two dimensions in region and time, Neng Gouyou
The degree of accuracy of the raising Similarity Measure of effect；Similarity is determined by the way of characteristic vector is calculated
Exponential complexity is relatively low, improves operation efficiency, reduces the performance requirement to arithmetic facility.
According to another aspect of the present invention, a kind of similar users identifying device is proposed, including：
Data extraction module, position and time for extracting the generation of user's communication behavior；Common index
Acquisition module, for according to the user position that communication behavior occurs within a predetermined period of time and time
Calculate base station common index of the user relative to base station；Characteristic vector acquisition module, for basis
User extracts predetermined quantity base station relative to the base station common index of different base station, and generation user is normal
Use base station characteristic vector；Index similarity determining module, it is normal for the user according to different user
User's index similarity is obtained with base station characteristic vector.
Alternatively, user's communication behavior includes conversing and/or connection data network；User communicates
The base station that the position that behavior occurs is interacted when occurring communication behavior for user.
Alternatively, common index acquisition module includes：Periodic index determining unit, for pre-
Fix time in section, obtain predetermined period base station common index, wherein, predetermined period base station is commonly used
Index is that user occurred to make a reservation in the number and predetermined amount of time of the predetermined period interacted with base station
The ratio of the number in cycle；Predetermined period includes one month, ten days, seven days, three days and/or one
Day；Common index determining unit, for determining that base station is normal according to predetermined period base station common index
Use index.
Alternatively, periodic index determining unit is used for：According to formula
Predetermined period base station common index is determined, wherein, I is predetermined period base station common index,
N is the number of predetermined period in predetermined amount of time, and i is the predetermined period mark in predetermined amount of time
Number, tiWhether occurred the mark interacted with base station for user in i-th of predetermined period：If
User occurred to interact with base station in i predetermined period, then tiFor 1；If i-th of predetermined week
User did not occurred to interact with base station in phase, then tiFor 0.
Alternatively, common index determining unit is used for：According to formula
CI=35My+30Ty+16Wy+12THy+8Dy
Base station common index is determined, wherein, CI is that the base station of user within a predetermined period of time is normal
With index, MyFor the predetermined period base station common index that predetermined period is one month；TyIt is predetermined
Cycle is the predetermined period base station common index of ten days；WyFor the predetermined week that predetermined period is seven
Phase base station common index；THyFor the predetermined period base station common index that predetermined period is three；
DyFor the predetermined period base station common index that predetermined period is one.
Alternatively, index similarity acquisition module is used for：According to formula
S=(Pm∩Pn)/Y
Determine user's index similarity, wherein, m, n be user mark, S be user n with
User m index similarity, PmFor user m conventional base station characteristic vector, PnFor user
N conventional base station characteristic vector, Pm∩PnFor PmWith PnMiddle identical base station number, Y is
User often uses the quantity of base station in the characteristic vector of base station.
Alternatively, data extraction module is used to extract the position that the communication behavior of user job day occurs
Put and the time；Common index acquisition module be used for according to user in predetermined amount of time it is workaday
The position and Time Calculation user that communication behavior occurs are commonly used relative to the working day base station of base station to be referred to
Number；Characteristic vector acquisition module is used for normal relative to the working day base station of different base station according to user
Predetermined quantity base station, the conventional base station characteristic vector of generation user job day are extracted with the size of index；
Index similarity determining module be used for according to day conventional base station feature of user job of different user to
Amount determines user's index similarity.
Alternatively, data extraction module is used to extract the position that the communication behavior of user's festivals or holidays occurs
Put and the time；Common index acquisition module be used for according to user in predetermined amount of time festivals or holidays
The position and Time Calculation user that communication behavior occurs are commonly used relative to the festivals or holidays base station of base station to be referred to
Number；Characteristic vector acquisition module is used for normal relative to the festivals or holidays base station of different base station according to user
Predetermined quantity base station is extracted with the size of index, generation user's festivals or holidays often use base station characteristic vector；
Index similarity determining module be used for according to user's festivals or holidays of different user often with base station feature to
Amount determines user's index similarity.
Alternatively, common index acquisition module be used for according to user in predetermined amount of time working day
Communication behavior occur position and Time Calculation user relative to base station working day base station commonly use
Index, and according to the communication behavior of user's festivals or holidays in predetermined amount of time occur position and
Festivals or holidays base station common index of the Time Calculation user relative to base station；Characteristic vector acquisition module
For extracting predetermined relative to the size of the working day base station common index of different base station according to user
Quantity base station, day conventional base station characteristic vector of generation user job, and according to user relative to
The size of the festivals or holidays base station common index of different base station extracts predetermined quantity base station, generates user
Festivals or holidays often use base station characteristic vector；Index similarity determining module includes：Working day similarity
Determining unit, determines to use for the conventional base station characteristic vector of user job day according to different user
Family index similarity；Festivals or holidays similarity determining unit, for being saved according to the user of different user
Holiday often determines user's index similarity with base station characteristic vector；Comprehensive similarity determining unit,
For determining that user is comprehensive based on user job day index similarity and user's festivals or holidays index similarity
Close index similarity.
Alternatively, in addition to：Similar users determining module, for by index similarity with it is predetermined
Threshold value compares, if index similarity is not less than predetermined threshold, it is determined that user is similar users；
If index similarity is less than predetermined threshold, it is determined that user is non-similar users.
The position and time that such device can be occurred based on communication behavior obtain user and commonly used
Base station characteristic vector, often user's index similarity is calculated further according to user with base station characteristic vector,
With good versatility；Due to consideration that region and time two dimensions, can be effective
Improve the degree of accuracy of Similarity Measure；Index similarity is determined by the way of characteristic vector is calculated
Complexity is relatively low, improves operation efficiency, reduces the performance requirement to arithmetic facility.
Brief description of the drawings
Accompanying drawing described herein is used for providing a further understanding of the present invention, constitutes the application
A part, schematic description and description of the invention is used to explain the present invention, not structure
Into inappropriate limitation of the present invention.In the accompanying drawings：
Fig. 1 is the flow chart of one embodiment of the similar users recognition methods of the present invention.
Fig. 2 is the flow chart of another embodiment of the similar users recognition methods of the present invention.
Fig. 3 is the flow chart of another embodiment of the similar users recognition methods of the present invention.
Fig. 4 is the schematic diagram of one embodiment of the similar users identifying device of the present invention.
Fig. 5 is the schematic diagram of another embodiment of the similar users identifying device of the present invention.
Fig. 6 is the schematic diagram of another embodiment of the similar users identifying device of the present invention.
Embodiment
Below by drawings and examples, technical scheme is done and further retouched in detail
State.
The flow chart of one embodiment of the similar users recognition methods of the present invention is as shown in Figure 1.
In a step 101, position and time that user's communication behavior occurs are extracted.In a reality
Apply in example, the communication behavior of user can include talk business, can also include connection data network
Network.The base station that the position that user's communication behavior occurs is interacted when can occur communication behavior with user
As mark, the date of communication behavior can occur for user for the time that user's communication behavior occurs.
In one embodiment, base station data can be extracted, determines what user interacted with the base station
Time.
In a step 102, according to user within a predetermined period of time communication behavior occur position and
Base station common index of the Time Calculation user relative to base station.The base station common index of user can be with
Referred to as user base station fingerprint index, embodies user and the base station is commonly used within a predetermined period of time
Degree.
In step 103, extract pre- relative to the base station common index of different base station according to user
Fixed number amount base station, generation user often uses base station characteristic vector.In one embodiment, can be by
User sorts relative to the base station common index of each base station according to order from big to small, extracts
The base station generation user of predetermined quantity often uses base station characteristic vector.
At step 104, user is often determined with base station characteristic vector according to the user of different user
Index similarity.The common factor of base station characteristic vector can be often used by calculating the user of two users
Mode determine the index similarity of two users.
By such method, the position and time that can be occurred based on communication behavior obtain user
Conventional base station characteristic vector, often calculates user's similarity with base station characteristic vector further according to user and refers to
Number, with good versatility；Due to consideration that two dimensions in region and time, can be effective
Raising Similarity Measure the degree of accuracy；Determine that similarity refers to by the way of characteristic vector is calculated
Number complexity is relatively low, improves operation efficiency, reduces the performance requirement to arithmetic facility.
Flow chart such as Fig. 2 institutes of another embodiment of the similar users recognition methods of the present invention
Show.
In step 201, position and time that user's communication behavior occurs are extracted.
In step 202., determine user within a predetermined period of time according to the predetermined period of setting
Predetermined period base station common index.Predetermined period can include one month, ten days, seven days, three
Day, one day.Predetermined period base station common index can be user and base station within a predetermined period of time
Occurred the ratio of the number of predetermined period in the number and predetermined amount of time of interactive predetermined period
Value.In one embodiment, can be according to formula：
Calculate predetermined period base station common index.Wherein, I is predetermined period base station common index,
N is the number of predetermined period in predetermined amount of time, and i is the predetermined period mark in predetermined amount of time
Number, tiWhether occurred the mark interacted with base station for user in i-th of cycle：If at i-th
User occurred to interact with base station in predetermined period, then ti=1；If in i-th of predetermined period
User did not occurred to interact with base station, then ti=0.
In step 203, determined according to the predetermined period base station common index of different predetermined periods
Base station common index.In one embodiment, can be according to formula：
CI=35My+30Ty+16Wy+12THy+8Dy (2)
Base station common index is determined, wherein, CI is that the base station of user within a predetermined period of time is normal
With index, MyFor the predetermined period base station common index that predetermined period is one month；TyIt is predetermined
Cycle is the predetermined period base station common index of ten days；WyFor the predetermined week that predetermined period is seven
Phase base station common index；THyFor the predetermined period base station common index that predetermined period is three；
DyFor the predetermined period base station common index that predetermined period is one.
In step 204, extract pre- relative to the base station common index of different base station according to user
Fixed number amount base station, generation user often uses base station characteristic vector.In one embodiment, can be with base
Occurs interactive base station generation user mutual region circle information, note within a predetermined period of time in user
Employ base station common index of the family relative to each base station.In one embodiment, it will can use
Family is sorted relative to the base station common index of each base station according to order from big to small, extracts pre-
The base station generation user of fixed number amount often uses base station characteristic vector.In one embodiment, Yong Huchang
Can be P=(C with base station characteristic vector1, C2, C3..., CY), wherein, C1、C2、
C3、CYIt is Base Station Identification, Y is predetermined quantity.
In step 205, user is often determined with base station characteristic vector according to the user of different user
Index similarity.The common factor of base station characteristic vector can be often used by calculating the user of two users
Mode determine the index similarity of two users.In one embodiment, can be according to formula
S=(Pm∩Pn)/Y (3)
Determine user's index similarity, wherein, m, n be user mark, S be user n with
User m index similarity, PmFor user m conventional base station characteristic vector, PnFor user
N conventional base station characteristic vector, Pm∩PnFor PmWith PnMiddle identical base station number, Y is
User often uses the quantity of base station in the characteristic vector of base station.
By such method, it can determine that user's similarity refers to by the calculating of low complex degree
Number, improves operation efficiency, reduces the requirement to operational outfit；It is normal with predetermined period base station
Base station common index is determined with index, the periodicity of user behavior has been fully taken into account, has made calculating
Result it is more accurate.
In one embodiment, can once it be calculated every predetermined period, it is determined that predetermined week
Phase base station common index simultaneously stores result of calculation, when reaching the deadline of predetermined amount of time,
Predetermined period base station common index according to being calculated in the predetermined amount of time determines the conventional finger in base station
Number, and base station common index is stored in feature database, calculated and used according to the data stored in feature database
Often use base station characteristic vector in family.Such as：Predetermined period includes one month, ten days, seven days, three days
With one day, then using the from date of predetermined amount of time as starting point, daily calculate be with one day once
The predetermined period base station common index in cycle, calculating in every three days are once with making a reservation for for the cycle on the three
Cycle base station common index, calculating in every seven days are once normal for the predetermined period base station in cycle with seven days
Calculated once with the predetermined period base station common index, every for the cycle on the ten with index, every ten days
The moon calculates predetermined period base station common index once by the cycle of January, is reaching the scheduled time
During the deadline of section, calculated according to all predetermined period base stations common index in predetermined amount of time
Base station common index.
By such method, can over time step by step calculation predetermined period base station it is normal
With index, the result of calculation before basis at the end of predetermined amount of time obtains base station common index,
So as to further reduce the stand-by period of computing, operation efficiency is improved.
In one embodiment, the position that the communication behavior of user on weekdays occurs can be extracted
And the time, based on the user position that workaday communication behavior occurs within a predetermined period of time and when
Between obtain is working day base station common index of the user relative to base station.In one embodiment,
Formula (1) can be utilized, the evaluation work day based on the working days evidence in predetermined amount of time
Base station common index.The conventional base station of user job day is obtained according to working day base station common index special
Vector is levied, is calculated and obtained based on the conventional base station characteristic vector of user job day of different user
User's index similarity be user job day index similarity, user can be embodied on weekdays
Similar situation, be easy to recognize working day similar users.
In one embodiment, the position that communication behavior of the user in festivals or holidays occurs can be extracted
And the time, based on user within a predetermined period of time festivals or holidays communication behavior occur position and when
Between obtain is festivals or holidays base station common index of the user relative to base station.In one embodiment,
Formula (1) can be utilized, festivals or holidays are calculated based on the festivals or holidays data in predetermined amount of time
Base station common index.User's festivals or holidays are obtained according to festivals or holidays base station common index often special with base station
Vector is levied, is often obtained by user's festivals or holidays of different user with being calculated based on the characteristic vector of base station
User's index similarity be user's festivals or holidays index similarity, user can be embodied in festivals or holidays
Similar situation, be easy to recognize festivals or holidays similar users.
Because festivals or holidays and workaday user behavior have a larger difference, thus by working day and
The difference of festivals or holidays, which accounts for scope, can realize the more accurately calculating to user's similarity；
Similarity Measure based on two dimensions of time and region can realize the similar use of high accuracy
Family is recognized.
In one embodiment, user job day index similarity and user's section can be obtained respectively
Holiday index similarity, coordinates corresponding predetermined weights to obtain user's comprehensive similarity index, base
Judge user's similar situation in user's comprehensive similarity index.In one embodiment, can be with base
In formula：
S=A*Sw+B*Sh
Calculate user's comprehensive similarity index.Wherein, S is user's comprehensive similarity index, Sw
For user job day index similarity, ShFor user's festivals or holidays index similarity, A is user's work
Make day index similarity weight, B is user's festivals or holidays index similarity weight.
By such method, it can be integrated based on user job day and the similar conditions of festivals or holidays
Consider user's similarity, so as to obtain more comprehensive user's index similarity, make similar users
Identification it is more accurate.
Flow chart such as Fig. 3 institutes of another embodiment of the similar users recognition methods of the present invention
Show.
In step 301, position and time that user's communication behavior occurs are extracted.In a reality
Apply in example, base station data can be extracted, determine the time that user interacts with the base station.
In step 302, according to user within a predetermined period of time communication behavior occur position and
Base station common index of the Time Calculation user relative to base station.
In step 303, extract pre- relative to the base station common index of different base station according to user
Fixed number amount base station, generation user often uses base station characteristic vector.In one embodiment, can be by
User sorts relative to the base station common index of each base station according to order from big to small, extracts
The base station generation user of predetermined quantity often uses base station characteristic vector.
In step 304, user is often determined with base station characteristic vector according to the user of different user
Index similarity.The common factor of base station characteristic vector can be often used by calculating the user of two users
Mode determine the index similarity of two users.
In step 305, user's index similarity is compared with predetermined threshold；If similarity
Index is not less than predetermined threshold, it is determined that two users are similar users；If index similarity is less than
Predetermined threshold, it is determined that two users are non-similar users.In one embodiment, it can export
Similar users inventory, or mark similar users, are easy to subsequent treatment and research.
By such method, the method that can be compared by threshold value refers to according to the similarity of user
Number determines similar users, so as to recognize similar users, is easy to be entered according to similar users data
Row is handled and researched and analysed.
One embodiment schematic diagram of the similar users identifying device of the present invention is as shown in Figure 4.Its
In, data extraction module 401 can extract position and the time of the generation of user's communication behavior.
In one embodiment, the communication behavior of user can include talk business, can also include connection
Data network.The position that user's communication behavior occurs is interacted when can occur communication behavior with user
Base station as mark, for user communication behavior can occur for the time that user's communication behavior occurs
Date.In one embodiment, base station data can be extracted, determines that user sends out with the base station
The time of raw interaction.Common index acquisition module 402 can be according to user within a predetermined period of time
Communication behavior occur position and Time Calculation user relative to base station base station common index.With
The base station common index at family is properly termed as user base station fingerprint index, embodies user in pre- timing
Between in section to the conventional degree of the base station.Characteristic vector acquisition module 403 can be according to user's phase
Base station common index for different base station extracts predetermined quantity base station, and generation user often uses base station
Characteristic vector.In one embodiment, user can be commonly used relative to the base station of each base station
Index sorts according to order from big to small, extracts the base station generation user Chang Yongji of predetermined quantity
Stand characteristic vector.Index similarity determining module 404 can be conventional according to the user of different user
Base station characteristic vector determines user's index similarity, in one embodiment, can be by calculating
The user of two users often determines that two users' is similar with the mode of the common factor of base station characteristic vector
Spend index.
The position and time that such device can be occurred based on communication behavior obtain user and commonly used
Base station characteristic vector, often user's index similarity is calculated further according to user with base station characteristic vector,
With good versatility；Due to consideration that region and time two dimensions, can be effective
Improve the degree of accuracy of Similarity Measure；Index similarity is determined by the way of characteristic vector is calculated
Complexity is relatively low, improves operation efficiency, reduces the performance requirement to arithmetic facility.
Another embodiment schematic diagram of the similar users identifying device of the present invention is as shown in Figure 5.
Wherein, data extraction module 51 is used for position and the time for extracting the generation of user's communication behavior.Often
Include periodic index determining unit 521 and common index determining unit with index acquisition module 52
522, periodic index determining unit 521 is used to first determine user pre- according to the predetermined period of setting
The predetermined period base station common index fixed time in section.Predetermined period can include one month, ten
Day, seven, three, one.Predetermined period base station common index can be in predetermined amount of time
With base station predetermined period in the number and predetermined amount of time of the predetermined period interacted occurred for interior user
Number ratio.In one embodiment, can be according to formula：
Calculate predetermined period base station common index.Wherein, I is predetermined period base station common index,
N is the number of predetermined period in predetermined amount of time, and i is the predetermined period mark in predetermined amount of time
Number, tiWhether occurred the mark interacted with base station for user in i-th of cycle：If at i-th
User occurred to interact with base station in predetermined period, then ti=1；If in i-th of predetermined period
User did not occurred to interact with base station, then ti=0.
Common index determining unit 522 is normal for the predetermined period base station according to different predetermined periods
Base station common index is determined with index.In one embodiment, can be according to formula：
CI=35My+30Ty+16Wy+12THy+8Dy (2)
Base station common index is determined, wherein, CI is that the base station of user within a predetermined period of time is normal
With index, MyFor the predetermined period base station common index that predetermined period is one month；TyIt is predetermined
Cycle is the predetermined period base station common index of ten days；WyFor the predetermined week that predetermined period is seven
Phase base station common index；THyFor the predetermined period base station common index that predetermined period is three；
DyFor the predetermined period base station common index that predetermined period is one.
Characteristic vector acquisition module 53 is used to be commonly used relative to the base station of different base station according to user
Exponent extracting predetermined quantity base station, generation user often uses base station characteristic vector.In one embodiment
In, base station common index that can be by user relative to each base station is according to order from big to small
Sequence, the base station generation user for extracting predetermined quantity often uses base station characteristic vector.In an implementation
In example, user often can be P=(C with base station characteristic vector1, C2, C3……CY), its
In, C1、C2、C3、CYIt is Base Station Identification, Y is predetermined quantity.
Index similarity determining module 54 is used to often use base station feature according to the user of different user
Vector determines user's index similarity.Can be often special with base station by calculating the user of two users
The mode for levying the common factor of vector determines the index similarity of two users.In one embodiment,
Can be according to formula
S=(Pm∩Pn)/Y (3)
Determine user's index similarity, wherein, m, n be user mark, S be user n with
User m index similarity, PmFor user m conventional base station characteristic vector, PnFor user
N conventional base station characteristic vector, Pm∩PnFor PmWith PnMiddle identical base station number, Y
The quantity of base station in the characteristic vector of base station is often used for user.
Such device can determine user's index similarity by the calculating of low complex degree, improve
Operation efficiency, reduces the requirement to equipment；Base is determined with predetermined period base station common index
Stand common index, fully taken into account the periodicity of user behavior, made the result of calculating more accurate
Really.
In one embodiment, periodic index determining unit 521 can be carried out every predetermined period
Once calculate, determine predetermined period base station common index and store result of calculation, it is predetermined when reaching
During the deadline of period, common index determining unit 522 is counted according in the predetermined amount of time
The predetermined period base station common index calculated determines base station common index, and by base station common index
Feature database is stored in, calculating user according to the data stored in feature database often uses base station characteristic vector.
Such as：Predetermined period includes one month, ten days, seven days, three days and one day, then with the scheduled time
The from date of section is starting point, and periodic index determining unit 521 is calculated daily was once with one day
The predetermined period base station common index in cycle, calculating in every three days are once with making a reservation for for the cycle on the three
Cycle base station common index, calculating in every seven days are once normal for the predetermined period base station in cycle with seven days
Calculated once with the predetermined period base station common index, every for the cycle on the ten with index, every ten days
The moon calculates predetermined period base station common index once by the cycle of January, is reaching the scheduled time
During the deadline of section, common index determining unit 522 is according to all predetermined in predetermined amount of time
Cycle base station common index calculation base station common index.
Such device can commonly use finger in step by step calculation predetermined period base station over time
Number, the result of calculation before basis at the end of predetermined amount of time obtains base station common index, from
And the stand-by period of computing is further reduced, improve operation efficiency.
In one embodiment, data extraction module is used to extract the communication row of user on weekdays
Position and time for generation, common index acquisition module are based on user's work within a predetermined period of time
What the position and time for making the communication behavior generation of day were obtained is working day of the user relative to base station
Base station common index, characteristic vector acquisition module obtains user according to working day base station common index
Working day often uses base station characteristic vector, and index similarity determining module is with user's work of different user
Make to commonly use day that to calculate obtained user's index similarity based on the characteristic vector of base station be user job
Day index similarity, can embody the similar situation of user on weekdays, be easy to recognize working day
Similar users.
In one embodiment, data extraction module is used to extract the communication row of user on weekdays
Position and time for generation, common index acquisition module are saved within a predetermined period of time based on user
What the position and time that the communication behavior of holiday occurs were obtained is festivals or holidays of the user relative to base station
Base station common index, characteristic vector acquisition module obtains user according to festivals or holidays base station common index
Festivals or holidays often use base station characteristic vector, and index similarity determining module is saved with the user of different user
User's index similarity that holiday is often obtained with being calculated based on the characteristic vector of base station is that user's section is false
Day index similarity, can embody similar situation of the user in festivals or holidays, be easy to recognize festivals or holidays
Similar users.
Because festivals or holidays and workaday user behavior have a larger difference, thus by working day and
The difference of festivals or holidays, which accounts for scope, can realize the more accurately calculating to user's similarity；
Similarity Measure based on two dimensions of time and region can realize the similar use of high accuracy
Family is recognized.
In one embodiment, index similarity determining module can be true including working day similarity
Order member, festivals or holidays similarity determining unit and comprehensive similarity determining unit, wherein, work
Day, similarity determining unit was used to obtain user job day index similarity, and festivals or holidays similarity is true
Order member is used to obtain user's festivals or holidays index similarity, and comprehensive similarity determining unit is used for base
Coordinate corresponding predetermined power in user job day index similarity and user's festivals or holidays index similarity
Value obtains user's comprehensive similarity index, judges that user is similar based on user's comprehensive similarity index
Situation.In one embodiment, comprehensive similarity determining unit can be based on formula：
S=A*Sw+B*Sh
Calculate user's comprehensive similarity index.Wherein, S is user's comprehensive similarity index, Sw
For user job day index similarity, ShFor user's festivals or holidays index similarity, A is user's work
Make day index similarity weight, B is user's festivals or holidays index similarity weight.
Such device can be considered based on the similar conditions of user job day and festivals or holidays
User's similarity, so as to obtain more comprehensive user's index similarity, makes the knowledge of similar users
Not it is more accurate.
Another embodiment schematic diagram of the similar users identifying device of the present invention is as shown in Figure 6.
Wherein, data extraction module 601, common index acquisition module 602, characteristic vector acquisition module
603 and the 26S Proteasome Structure and Function of index similarity determining module 604 it is similar to Fig. 4 embodiment.
Similar users identifying device also includes similar users determining module 605, for by user's similarity
Index is compared with predetermined threshold；If index similarity is not less than predetermined threshold, it is determined that dual-purpose
Family is similar users；If index similarity is less than predetermined threshold, it is determined that two users are non-similar
User.In one embodiment, similar users inventory, or mark similar users can be exported,
It is easy to subsequent treatment and research.
The method that such device can be compared by threshold value is true according to the index similarity of user
Determine similar users, so as to recognize similar users, be easy to according at similar users data
Manage and research and analyse.
Finally it should be noted that：The above embodiments are merely illustrative of the technical solutions of the present invention and
It is non-that it is limited；It is affiliated although the present invention is described in detail with reference to preferred embodiments
The those of ordinary skill in field should be understood：The embodiment of the present invention can still be entered
Row modification carries out equivalent substitution to some technical characteristics；Without departing from technical solution of the present invention
Spirit, it all should cover among claimed technical scheme scope of the invention.
Claims (14)
1. a kind of similar users recognition methods, it is characterised in that including：
Extract position and time that user's communication behavior occurs；
According to the user position that communication behavior occurs within a predetermined period of time and Time Calculation user
Relative to the base station common index of base station；
Predetermined quantity base is extracted relative to the base station common index of different base station according to user
Stand, generation user often uses base station characteristic vector；
Often determine that user's similarity refers to base station characteristic vector according to the user of different user
Number.
2. according to the method described in claim 1, it is characterised in that
User's communication behavior includes conversing and/or connection data network；
The base that the position that user's communication behavior occurs is interacted when occurring communication behavior for user
Stand.
3. according to the method described in claim 1, it is characterised in that described to be existed according to user
In predetermined amount of time communication behavior occur position and Time Calculation user relative to base station base station
Common index includes：
Predetermined period base station common index is obtained in the predetermined amount of time, wherein, it is described pre-
Fixed cycle base station common index is that the number of the predetermined period interacted occurred with base station for user
With the ratio of the number of the predetermined period in the predetermined amount of time；The predetermined period includes
One month, ten days, seven days, three days and/or one day；
The base station common index is determined according to predetermined period base station common index.
4. method according to claim 3, it is characterised in that the acquisition predetermined week
Phase base station common index includes：
According to formula
<mrow>
<mi>I</mi>
<mo>=</mo>
<mfrac>
<mrow>
<munder>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
<mo>,</mo>
<mn>2</mn>
<mo>,</mo>
<mn>3..</mn>
</mrow>
</munder>
<msub>
<mi>t</mi>
<mi>i</mi>
</msub>
</mrow>
<mi>n</mi>
</mfrac>
</mrow>
Predetermined period base station common index is determined, wherein, I is the predetermined period base station
Common index, n is the number of the predetermined period in the predetermined amount of time, and i is described predetermined
Predetermined period label in period, tiDescribed in i-th in predetermined period user whether with base station
Occurred interactive mark：If friendship occurred for user and base station in predetermined period described in i-th
Mutually, then tiFor 1；If user did not occurred to interact with base station in predetermined period described in i-th,
Then tiFor 0；
And/or,
It is described that the base station common index bag is determined according to predetermined period base station common index
Include：
According to formula
CI=35My+30Ty+16Wy+12THy+8Dy
The base station common index is determined, wherein, CI is user in the predetermined amount of time
The base station common index, MyFor the predetermined period base that the predetermined period is one month
Stand common index；TyFor the conventional finger in the predetermined period base station that the predetermined period is ten
Number；WyFor the predetermined period base station common index that the predetermined period is seven；THy
For the predetermined period base station common index that the predetermined period is three；DyMake a reservation for be described
Cycle is the predetermined period base station common index of one day.
5. according to the method described in claim 1, it is characterised in that described to be used according to different
The user at family often determines that user's index similarity includes with base station characteristic vector：
According to formula
S=(Pm∩Pn)/Y
User's index similarity is determined, wherein, m, n identify for user, and S is user n
With the user m index similarity, PmBase station feature is often used for the user m user
Vector, PnBase station characteristic vector, P are often used for the user n userm∩PnFor PmWith Pn
Middle identical base station number, Y is the quantity that the user often uses base station in the characteristic vector of base station.
6. according to the method described in claim 1, it is characterised in that：
The position for extracting the generation of user's communication behavior and time include：Extract user job day
Communication behavior occur position and the time；
It is described according to user within a predetermined period of time communication behavior occur position and Time Calculation
User includes relative to the base station common index of base station：Worked according to user in predetermined amount of time
The position and Time Calculation user that the communication behavior of day occurs are normal relative to the working day base station of base station
Use index；
It is described that predetermined number is extracted relative to the base station common index of different base station according to user
Base station is measured, generation user is often included with base station characteristic vector：According to user relative to different base station
The size of working day base station common index extract predetermined quantity base station, generate user job
Day conventional base station characteristic vector；
The user according to different user often determines that user is similar with base station characteristic vector
Degree index includes：Determined according to the conventional base station characteristic vector of the user job of different user day
The user job day index similarity；
And/or,
The position for extracting the generation of user's communication behavior and time include：Extract user's festivals or holidays
Communication behavior occur position and the time；
It is described according to user within a predetermined period of time communication behavior occur position and Time Calculation
User includes relative to the base station common index of base station：Vacation is saved in predetermined amount of time according to user
The position and Time Calculation user that the communication behavior of day occurs are normal relative to the festivals or holidays base station of base station
Use index；
It is described that predetermined number is extracted relative to the base station common index of different base station according to user
Base station is measured, generation user is often included with base station characteristic vector：According to user relative to different base station
The size of festivals or holidays base station common index extract predetermined quantity base station, generation user's section is false
Day conventional base station characteristic vector；
The user according to different user often determines that user is similar with base station characteristic vector
Degree index includes：Often determined according to user's festivals or holidays of different user with base station characteristic vector
User's festivals or holidays index similarity；
And/or,
It is described according to user within a predetermined period of time communication behavior occur position and Time Calculation
User includes relative to the base station common index of base station：Worked according to user in predetermined amount of time
The position and Time Calculation user that the communication behavior of day occurs are normal relative to the working day base station of base station
Use index；According to the communication behavior of user's festivals or holidays in predetermined amount of time occur position and when
Between calculate user relative to base station festivals or holidays base station common index；
It is described that predetermined number is extracted relative to the base station common index of different base station according to user
Base station is measured, generation user is often included with base station characteristic vector：According to user relative to different base station
The size of working day base station common index extract predetermined quantity base station, generate user job
Day conventional base station characteristic vector；It is normal relative to the festivals or holidays base station of different base station according to user
Predetermined quantity base station is extracted with the size of index, generation user's festivals or holidays often use base station characteristic vector；
The user according to different user often determines that user is similar with base station characteristic vector
Degree index includes：Determined according to the conventional base station characteristic vector of the user job of different user day
The user job day index similarity；Often base is used according to user's festivals or holidays of different user
Characteristic vector of standing determines user's festivals or holidays index similarity；Based on the user job day phase
User's comprehensive similarity index is determined like degree index and user's festivals or holidays index similarity.
7. according to any described method of claim 1~6, it is characterised in that also include：
The index similarity is compared with predetermined threshold；
If the index similarity is not less than the predetermined threshold, it is determined that user uses to be similar
Family；
If the index similarity is less than the predetermined threshold, it is determined that user is non-similar use
Family.
8. a kind of similar users identifying device, it is characterised in that including：
Data extraction module, position and time for extracting the generation of user's communication behavior；
Common index acquisition module, for communication behavior to occur within a predetermined period of time according to user
Position and Time Calculation user relative to base station base station common index；
Characteristic vector acquisition module, refers to for being commonly used according to user relative to the base station of different base station
Number extracts predetermined quantity base station, and generation user often uses base station characteristic vector；
Index similarity determining module, for often special with base station according to the user of different user
Levy vector and obtain user's index similarity.
9. device according to claim 8, it is characterised in that
User's communication behavior includes conversing and/or connection data network；
The base that the position that user's communication behavior occurs is interacted when occurring communication behavior for user
Stand.
10. device according to claim 8, it is characterised in that the common index is obtained
Modulus block includes：
Periodic index determining unit, for obtaining predetermined period base station in the predetermined amount of time
Common index, wherein, predetermined period base station common index is that with base station friendship occurred for user
The number of the mutual predetermined period and the number of the predetermined period in the predetermined amount of time
Ratio；The predetermined period includes one month, ten days, seven days, three days and/or one day；
Common index determining unit, for determining institute according to predetermined period base station common index
State base station common index.
11. device according to claim 10, it is characterised in that the periodic index
Determining unit is used for：
According to formula
<mrow>
<mi>I</mi>
<mo>=</mo>
<mfrac>
<mrow>
<munder>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
<mo>,</mo>
<mn>2</mn>
<mo>,</mo>
<mn>3..</mn>
</mrow>
</munder>
<msub>
<mi>t</mi>
<mi>i</mi>
</msub>
</mrow>
<mi>n</mi>
</mfrac>
</mrow>
Predetermined period base station common index is determined, wherein, I is the predetermined period base station
Common index, n is the number of the predetermined period in the predetermined amount of time, and i is described predetermined
Predetermined period label in period, tiDescribed in i-th in predetermined period user whether with base station
Occurred interactive mark：If friendship occurred for user and base station in predetermined period described in i-th
Mutually, then tiFor 1；If user did not occurred to interact with base station in predetermined period described in i-th,
Then tiFor 0；
And/or,
The common index determining unit is used for：
According to formula
CI=35My+30Ty+16Wy+12THy+8Dy
The base station common index is determined, wherein, CI is user in the predetermined amount of time
The base station common index, MyFor the predetermined period base that the predetermined period is one month
Stand common index；TyFor the conventional finger in the predetermined period base station that the predetermined period is ten
Number；WyFor the predetermined period base station common index that the predetermined period is seven；THy
For the predetermined period base station common index that the predetermined period is three；DyMake a reservation for be described
Cycle is the predetermined period base station common index of one day.
12. device according to claim 8, it is characterised in that the index similarity
Acquisition module is used for：
According to formula
S=(Pm∩Pn)/Y
User's index similarity is determined, wherein, m, n identify for user, and S is user n
With the user m index similarity, PmBase station feature is often used for the user m user
Vector, PnBase station characteristic vector, P are often used for the user n userm∩PnFor PmWith Pn
Middle identical base station number, Y is the quantity that the user often uses base station in the characteristic vector of base station.
13. device according to claim 8, it is characterised in that：
The data extraction module is used to extract the position that the communication behavior of user job day occurs
And the time；
The common index acquisition module is used to be worked in the predetermined amount of time according to user
The position and Time Calculation user that the communication behavior of day occurs are normal relative to the working day base station of base station
Use index；
The characteristic vector acquisition module is used for the work according to user relative to different base station
The size for making day base station common index extracts predetermined quantity base station, the conventional base of generation user job day
Stand characteristic vector；
The index similarity determining module is used for the user job day according to different user
Conventional base station characteristic vector determines user's index similarity；
And/or,
The data extraction module is used to extract the position that the communication behavior of user's festivals or holidays occurs
And the time；
The common index acquisition module is used to save vacation in the predetermined amount of time according to user
The position and Time Calculation user that the communication behavior of day occurs are normal relative to the festivals or holidays base station of base station
Use index；
The characteristic vector acquisition module is used for the section according to user relative to different base station
The size of holiday base station common index extracts predetermined quantity base station, and generation user's festivals or holidays often use base
Stand characteristic vector；
The index similarity determining module is used for user's festivals or holidays according to different user
Conventional base station characteristic vector determines user's index similarity；
And/or,
The common index acquisition module is used to be worked in the predetermined amount of time according to user
The position and Time Calculation user that the communication behavior of day occurs are normal relative to the working day base station of base station
With index, and occurred according to the communication behavior of user's festivals or holidays in the predetermined amount of time
The festivals or holidays base station common index of position and Time Calculation user relative to base station；
The characteristic vector acquisition module is used for the work according to user relative to different base station
The size for making day base station common index extracts predetermined quantity base station, the conventional base of generation user job day
Stand characteristic vector, and commonly used and refer to relative to the festivals or holidays base station of different base station according to user
Several sizes extracts predetermined quantity base station, and generation user's festivals or holidays often use base station characteristic vector；
The index similarity determining module includes：
Working day similarity determining unit, for the user job according to different user
Day, conventional base station characteristic vector determined user's index similarity；
Festivals or holidays similarity determining unit, it is false for being saved according to the user of different user
Day, conventional base station characteristic vector determined user's index similarity；
Comprehensive similarity determining unit, for based on the user job day index similarity
User's comprehensive similarity index is determined with user's festivals or holidays index similarity.
14. according to any described device of claim 8~13, it is characterised in that also include：
Similar users determining module, for the index similarity to be compared with predetermined threshold,
If the index similarity is not less than the predetermined threshold, it is determined that user is similar users；If
The index similarity is less than the predetermined threshold, it is determined that user is non-similar users.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610268871.0A CN107316044A (en) | 2016-04-27 | 2016-04-27 | Similar users recognition methods and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610268871.0A CN107316044A (en) | 2016-04-27 | 2016-04-27 | Similar users recognition methods and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107316044A true CN107316044A (en) | 2017-11-03 |
Family
ID=60184575
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610268871.0A Pending CN107316044A (en) | 2016-04-27 | 2016-04-27 | Similar users recognition methods and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107316044A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108318748A (en) * | 2018-02-02 | 2018-07-24 | 湘潭大学 | A kind of base station electromagnetic radiation intensity similarity estimating method |
| CN109769210A (en) * | 2018-11-23 | 2019-05-17 | 亚信科技（中国）有限公司 | User Activity Regional Similarity judgment method, device, computer equipment |
| CN110856159A (en) * | 2018-08-21 | 2020-02-28 | 中国移动通信集团湖南有限公司 | Method, device and storage medium for determining family circle members |
| WO2020055321A1 (en) * | 2018-09-10 | 2020-03-19 | Eureka Analytics Pte. Ltd. | Telecommunications data used for lookalike analysis |
| CN111884821A (en) * | 2020-03-27 | 2020-11-03 | 马洪涛 | Ticket data processing and displaying method and device and electronic equipment |
| CN112738724A (en) * | 2020-12-17 | 2021-04-30 | 福建新大陆软件工程有限公司 | Method, device, equipment and medium for accurately identifying regional target crowd |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101924609A (en) * | 2009-06-12 | 2010-12-22 | 中国移动通信集团公司 | Processing method of uplink data in coordinated multipoint transmission and related device |
| CN101945400A (en) * | 2009-07-10 | 2011-01-12 | 中国移动通信集团公司 | User dynamic behavior analysis method and analysis device |
| CN103700018A (en) * | 2013-12-16 | 2014-04-02 | 华中科技大学 | Method for dividing users in mobile social network |
| US20150172859A1 (en) * | 2010-02-22 | 2015-06-18 | Nokia Technologies Oy | Accurate gnss time handling in dual/multi-sim terminals |
| CN104881459A (en) * | 2015-05-22 | 2015-09-02 | 电子科技大学 | Friend recommendation method of mobile social network |
-
2016
- 2016-04-27 CN CN201610268871.0A patent/CN107316044A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101924609A (en) * | 2009-06-12 | 2010-12-22 | 中国移动通信集团公司 | Processing method of uplink data in coordinated multipoint transmission and related device |
| CN101945400A (en) * | 2009-07-10 | 2011-01-12 | 中国移动通信集团公司 | User dynamic behavior analysis method and analysis device |
| US20150172859A1 (en) * | 2010-02-22 | 2015-06-18 | Nokia Technologies Oy | Accurate gnss time handling in dual/multi-sim terminals |
| CN103700018A (en) * | 2013-12-16 | 2014-04-02 | 华中科技大学 | Method for dividing users in mobile social network |
| CN104881459A (en) * | 2015-05-22 | 2015-09-02 | 电子科技大学 | Friend recommendation method of mobile social network |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108318748A (en) * | 2018-02-02 | 2018-07-24 | 湘潭大学 | A kind of base station electromagnetic radiation intensity similarity estimating method |
| CN108318748B (en) * | 2018-02-02 | 2020-06-12 | 湘潭大学 | Method for evaluating similarity of electromagnetic radiation intensity of base station |
| CN110856159A (en) * | 2018-08-21 | 2020-02-28 | 中国移动通信集团湖南有限公司 | Method, device and storage medium for determining family circle members |
| CN110856159B (en) * | 2018-08-21 | 2022-07-26 | 中国移动通信集团湖南有限公司 | Method, device and storage medium for determining family circle members |
| WO2020055321A1 (en) * | 2018-09-10 | 2020-03-19 | Eureka Analytics Pte. Ltd. | Telecommunications data used for lookalike analysis |
| CN109769210A (en) * | 2018-11-23 | 2019-05-17 | 亚信科技（中国）有限公司 | User Activity Regional Similarity judgment method, device, computer equipment |
| CN111884821A (en) * | 2020-03-27 | 2020-11-03 | 马洪涛 | Ticket data processing and displaying method and device and electronic equipment |
| CN111884821B (en) * | 2020-03-27 | 2022-04-29 | 马洪涛 | Ticket data processing and displaying method and device and electronic equipment |
| CN112738724A (en) * | 2020-12-17 | 2021-04-30 | 福建新大陆软件工程有限公司 | Method, device, equipment and medium for accurately identifying regional target crowd |
| CN112738724B (en) * | 2020-12-17 | 2022-09-23 | 福建新大陆软件工程有限公司 | Method, device, equipment and medium for accurately identifying regional target crowd |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107316044A (en) | Similar users recognition methods and device | |
| CN110197331A (en) | Business data processing method, device, equipment and computer readable storage medium | |
| CN105281925B (en) | The method and apparatus that network service groups of users divides | |
| CN110032954A (en) | A kind of reinforcing bar intelligent recognition and method of counting and system | |
| CN109685092B (en) | Clustering method, equipment, storage medium and device based on big data | |
| CN106021329A (en) | A user similarity-based sparse data collaborative filtering recommendation method | |
| CN109711801A (en) | A kind of Internetbank account checking method and device | |
| CN103927309A (en) | Method and device for marking information labels for business objects | |
| CN112650743B (en) | Funnel data analysis method, system, electronic equipment and storage medium | |
| CN108280183B (en) | Information pushing system based on big data matching and GPS positioning | |
| CN109918645A (en) | Method, apparatus, computer equipment and the storage medium of depth analysis text | |
| CN110232131A (en) | Intention material searching method and device based on intention label | |
| CN111092764B (en) | Real-time dynamic affinity relation analysis method and system | |
| CN110209674A (en) | A kind of the cloud database statistical method and device of industrial environment dust control wechat small routine | |
| CN111652486A (en) | Method and device for calculating service index | |
| CN107181601B (en) | Flow reminding method and device | |
| CN110766441A (en) | Resource object processing method and device, storage medium and computer equipment | |
| CN114358979A (en) | Hotel matching method and device, electronic equipment and storage medium | |
| CN109540002A (en) | Cargo compartment dimension measurement method and device | |
| CN110532863A (en) | Gesture operation method, device and computer equipment | |
| CN114328785A (en) | Method and device for extracting road information | |
| CN109558462A (en) | Data statistical approach and device | |
| CN109784634A (en) | Coverage division methods, electronic device and readable storage medium storing program for executing | |
| CN106791230A (en) | Telephone number recognition methods and device | |
| CN110020123A (en) | A kind of promotion message put-on method, device, medium and equipment |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171103 |