CN104598452B - User's gender analysis method and apparatus - Google Patents
User's gender analysis method and apparatus Download PDFInfo
- Publication number
- CN104598452B CN104598452B CN201310526980.4A CN201310526980A CN104598452B CN 104598452 B CN104598452 B CN 104598452B CN 201310526980 A CN201310526980 A CN 201310526980A CN 104598452 B CN104598452 B CN 104598452B
- Authority
- CN
- China
- Prior art keywords
- user
- gender
- cis
- male
- probability
- 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.)
- Active
Links
Classifications
-
- 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
- G06F16/958—Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention provides a kind of user's gender analysis method and apparatus.It is related to data analysis field；Solve the problems, such as that existing analysis mode is not suitable for individual character domain name and is associated with weaker occasion with name.This method includes：Collecting sample data set, the sample data set include multipair user personality domain name and corresponding user's gender；Count the probability that different monograms occur according to gender on different alphabetical and adjacent several cis-positions on each cis-position in the user personality domain name that the sample data is concentrated；Using the ratio of sample data concentration male and the probability as parameter is referred to, the user personality domain name of unknown subscriber's gender is analyzed, judges user's gender.Technical solution provided by the invention is suitable for data analysis, realizes user's gender analysis based on automation algorithm.
Description
Technical field
The present invention relates to data analysis field more particularly to a kind of user's gender analysis method and apparatus.
Background technology
Under internet environment, the gender of user is a highly important information.According to the gender of user, in internet
Different contents can be showed to different user by holding supplier.For example, male user may compare female user to e-sports
It is interested, and female user may be interested in fashion dress ornament compared to male user.In this case, if user
Gender is identified that Internet advertising provider can be female user exhibition to show the advertisement of e-sports for male user
Show the advertisement of fashion dress ornament, so that advertisement is more targeted, obtains better advertising results.
For the user of registration blog, microblogging or other social network sites, many service providers can be complete in user
After necessary log-on message, it is proposed that user fills in the attribute of some users itself, such as gender, age, working condition, for certainly
Oneself setting individual character domain name etc., and be often usually all selectivity in the information registering item for being related to privacy of user in these attributes
Item is filled in, rather than has to fill out item, in this way, having led to quite a few user selection does not fill in this type of information, such as
User is that the information of oneself is protected not leak outside, and can select not fill in the age, gender etc., then, for data analysis mechanism or confession
It answers for quotient itself, also can not just directly acquire the gender information of user.But the selectivity for not being related to privacy fills in item
For, the success rate being filled out is often very high.For example, individual character domain name, service provider in order to increase user experience and affinity,
Often allow user that the virtual url for having and representing user's nature is set for the microblogging or personal space homepage of oneself.User
It can set these Domain Name Form registering sites to the name of oneself, or the arbitrary number oneself liked or monogram, i.e. fashion and side
Just.However, for the gender differences of mankind itself, in the setting to individual character domain name, male and female often go to set by instinct
Set some domain names for representing self attributes.For example, certain user may register property domain name one by one： http://weibo.com/
Basketballfans, wherein weibo.com are the domain names of microblogging service provider, and the parts basketballfans are user
The individual character domain name of selection.So, the gender information of user is extrapolated by the individual character domain name with user representative's property, i.e., do not invaded
Violate user and collects user information again.
In existing technology, most like technology is United States Patent (USP) 7,447,996 [1].This patent proposes one kind
Software module, the gender for inferring user according to different user names in instant communicating system, according to different gender exhibitions
Show different virtual images.Dependent on specific Praxeology data, i.e. name in language-specific and the pass between gender
System.For example, referred in this patent, for Chinese Name, by the retrieval of Praxeology database, " xiuxiu " and
" lili " it is more likely that women name.
Praxeology database is not particularly suited for multiple network application scenarios, is particularly unsuitable for individual character domain name and name
The weaker occasion of association.The composition of individual character domain name normally comprises a large amount of ingredients beyond common name scope, these ingredients are very
Difficulty passes through Praxeology data analysis.For example, may include " basketball ", i.e. basketball in individual character domain name；And it may incite somebody to action
Basketball is put into the basket ball fan of individual character domain name, and male may occupy an leading position.If by " basketball corresponds to male " this kind of number
According to database is added, required work will be significantly greatly increased, and be difficult complete.
Invention content
The present invention provides a kind of user's gender analysis method and apparatus, solve existing analysis mode and are not suitable for individual character
Domain name is associated with the problem of weaker occasion with name.
A kind of user's gender analysis method, including：
Collecting sample data set, the sample data set include multipair user personality domain name and corresponding user's gender；
It counts in the user personality domain name that the sample data is concentrated on each cis-position on different letters and adjacent several cis-positions
The probability that different monograms occur according to gender；
Using the ratio of sample data concentration male and the probability as parameter is referred to, to the use of unknown subscriber's gender
Family individual character domain name is analyzed, and judges user's gender.
Preferably, different alphabetical and adjacent on each cis-position in the user personality domain name that the statistics sample data is concentrated
Before the step of probability that different monograms occur according to gender on several cis-positions, further include：
Calculate the ratio that the sample data concentrates male.
Preferably, count different alphabetical and adjacent several on each cis-position in the user personality domain name that the sample data is concentrated
Monogram includes according to the probability that gender occurs on cis-position：
Step a：The part that user specifies in a user personality domain name is taken, while it is corresponding to record the user personality domain name
User's gender；
Step b：To different on the number and/or adjacent several cis-positions of letter appearance on each cis-position of the specified part
The number that monogram occurs is counted；
Step c：Processing such as step a to b is carried out to whole user personality domain names that the sample data is concentrated, until institute
Sample data set traversal is stated to complete；
Step d：Count number that different sexes occurs in letter on each cis-position of user personality domain name and/or adjacent
The number that different sexes occurs in monogram on several cis-positions, and calculate alphabetical and/or adjacent several suitable on each cis-position
The probability that different sexes occurs in monogram on position.
Preferably, the number and/or phase that different sexes occurs in letter on each cis-position of user personality domain name are counted
The number that different sexes occurs in monogram on adjacent several cis-positions, and calculate alphabetical and/or adjacent several on each cis-position
The probability that different sexes occurs in monogram on cis-position is specially：
According to expression formula
Calculate separately the probability that each letter on each cis-position corresponds to male with each monogram on adjacent several cis-positions.Its
In, the P (n-gram corresponds to male) on the left of equation is that length is that monogram on adjacent several cis-positions of n corresponds to male's
Probability, P (n-gram corresponds to male) corresponds to the probability of male for the letter on single cis-position when n is 1；N- on the right side of equation
Gram corresponds to that male's frequency is letter on single cis-position or length is that monogram on adjacent several cis-positions of n corresponds to man
The number of property, n-gram correspond to that women frequency is letter on single cis-position or length is letter on adjacent several cis-positions of n
Combination corresponds to the number of women.
Preferably, using the probability as parameter is referred to, the user personality domain name of unknown subscriber's gender is analyzed, is sentenced
Break user's gender include：
Step a：The length for obtaining the user personality domain name of unknown subscriber's gender, is denoted as k；
Step b：According to expression formula
The gender for calculating the user is the probability of male, wherein url indicates the part that user specifies in individual character domain name；
substr（Url, j, i）Indicate that jth position character starts what the monogram on adjacent several cis-positions that length is i was constituted in url
Substring is the substring that the letter on single cis-position is constituted when i is 1；N indicates the number of substr (url, j, i)；wh
Indicate the weight of the letter or monogram；P (substr (url, j, i) concentrates corresponding male in sample data) indicates above-mentioned
The corresponding male's probability of letter or monogram on substring；
Step c：Result of calculation and the sample data in comparison step b concentrate the ratio of male；
Step d：When result of calculation in stepb is more than or equal to the ratio that step c is calculated, the non-intellectual is judged
The gender of other user is male.
Preferably, after the step d, further include：
Step e：When result of calculation in stepb is less than the ratio that step c is calculated, the judgement unknown gender is used
The gender at family is women.
The present invention also provides a kind of user's gender analysis devices, including：
Sampling module is used for collecting sample data set, and the sample data set includes multipair user personality domain name and correspondence
User's gender；
Reference parameter computing module, it is different on each cis-position for counting in the user personality domain name that the sample data is concentrated
The probability that different monograms occur according to gender on alphabetical and adjacent several cis-positions；
Analysis module, the ratio and the probability for concentrating male using the sample data are as parameter is referred to, to not
Know that the user personality domain name of user's gender is analyzed, judges user's gender.
Preferably, which further includes：
With reference to ratio computing module, the ratio of male is concentrated for calculating the sample data.
Preferably, the reference parameter computing module includes：
Gender extraction unit for taking the part that user specifies in a user personality domain name, while recording the user
The property corresponding user's gender of domain name；
Counting unit, for letter occurs on each cis-position of the specified part number and/or adjacent several suitable
The number that different monograms occur on position is counted；
Statistic unit, the place for whole user personality domain names that the sample data is concentrated to be carried out with the counting unit
Reason counts letter on each cis-position of user personality domain name and different sexes is gone out until sample data set traversal is completed
The number that different sexes occurs in monogram on existing number and/or adjacent several cis-positions, and calculate word on each cis-position
The probability that different sexes occurs in monogram on female and/or adjacent several cis-positions.
Preferably, the statistic unit calculates on each cis-position on letter and/or adjacent several cis-positions monogram for not
With gender occur probability be specially：
According to expression formula
Calculate separately each letter and the corresponding male's probability of each monogram on adjacent several cis-positions on each cis-position, wherein P
(n-gram corresponds to male) is that length is the probability that a monogram corresponds to male on adjacent several cis-positions of n, P when n is 1
(n-gram corresponds to male) is the probability that a letter corresponds to male on single cis-position, and it is single suitable that n-gram, which corresponds to male's frequency,
A monogram corresponds to the number of male on the upper letter in position or adjacent several cis-positions that length is n, and n-gram corresponds to women
Frequency is the number that a letter or length are a monogram corresponds to women on adjacent several cis-positions of n on single cis-position.
Preferably, the analysis module includes：
Domain name length acquiring unit, the length of the user personality domain name for obtaining unknown subscriber's gender, is denoted as k；
Probability calculation unit, for according to expression formula
The gender for calculating the user is the probability of male, wherein url indicates the part that user specifies in individual character domain name,
substr（Url, j, i）Indicate that jth position character starts the substring that the adjacent character that length is i is constituted in url, n is indicated
The number of substr (url, j, i), whIndicate the weight of the letter or monogram, (substr (url, j, i) is in sample number by P
Male is corresponded to according to concentrating) indicate that jth position character or jth position character start the sub- character that the adjacent character that length is i is constituted in url
The corresponding male's probability of letter or monogram on string；
Comparing unit, the ratio that result of calculation and reference ratio computing module for comparing probability calculation unit are calculated
Example；
Judging unit, for being to compare the result of calculation of probability calculation unit to be more than in the comparing unit result of the comparison
When equal to the ratio being calculated with reference to ratio computing module, judge that the gender of the unknown gender user is male.
Preferably, the judging unit is additionally operable in the comparing unit result of the comparison be to compare probability calculation unit
Result of calculation when being less than the ratio being calculated with reference to ratio computing module, judge the gender of the unknown gender user
For women.
The present invention provides a kind of user's gender analysis method and apparatus, collecting sample data set, the sample data sets
Including multipair user personality domain name and corresponding user's gender, then count in the user personality domain name that the sample data is concentrated
The different letters probability that different monograms occur according to gender on adjacent several cis-positions on each cis-position, then made with the probability
For reference parameter, the user personality domain name of unknown subscriber's gender is analyzed, judges user's gender, is realized based on certainly
User's gender analysis of dynamicization algorithm, it is more flexible and accurate, it solves existing analysis mode and is not suitable for individual character domain first name and last name
Name is associated with the problem of weaker occasion.
Description of the drawings
Fig. 1 is a kind of flow chart for user's gender analysis method that the embodiment of the present invention one provides；
Fig. 2 is a kind of structural schematic diagram for user's gender analysis device that the embodiment of the present invention two provides；
Fig. 3 is the structural schematic diagram of reference parameter computing module 202 in Fig. 2；
Fig. 4 is the structural schematic diagram of analysis module 203 in Fig. 2.
Specific implementation mode
The embodiment provides a kind of user's gender analysis method and apparatus, by a kind of algorithm of automation,
Avoid the dependence to Praxeology database.
The embodiment of the present invention is described in detail below in conjunction with attached drawing.It should be noted that not conflicting
In the case of, the features in the embodiments and the embodiments of the present application mutually can be combined arbitrarily.
First in conjunction with attached drawing, the embodiment of the present invention one is illustrated.
An embodiment of the present invention provides a kind of user's gender analysis methods, and the stream of user's gender analysis is completed using this method
Journey is as shown in Figure 1, include：
Step 101, collecting sample data set, the sample data set include multipair user personality domain name and corresponding user
Gender；
Step 102 calculates the ratio that the sample data concentrates male；
Step 103, the statistics sample data are concentrated different on each cis-position in male's proportion and user personality domain name
The probability that different monograms occur according to gender on alphabetical and adjacent several cis-positions；
This step specifically includes：
Step a：The part that user specifies in a user personality domain name is taken, while it is corresponding to record the user personality domain name
User's gender；
Step b：Calculate the ratio that the sample data concentrates male；Step c：To on each cis-position of the specified part
The number that different monograms occur on the number and/or adjacent several cis-positions that letter occurs is counted；
The method counted for the number that letter occurs on single character bit is as follows：
The occurrence number of letter in the user personality domain name first is added 1, then again by the user personality domain
The occurrence number for the character string that name second is constituted adds 1, is counted successively to last position of the user personality domain name.
The method counted for the number that different monograms occur on adjacent several cis-positions is as follows：
The length n for the character string that adjacent several cis-positions are constituted is determined first, then with the user personality domain name first
It takes n cis-positions to constitute character string for starting, the number that the monogram in the character string occurs is added 1；Then with the user personality
Domain name second is that starting takes n cis-positions to constitute character string, and the number that the monogram in the character string occurs is added 1.Class according to this
It pushes away, until straight last position for causing the last position of character string as user personality domain name.The value of n is by 2 to the user personality domain name
Length.
Step c：Processing such as step a to b is carried out to whole user personality domain names that the sample data is concentrated, until institute
Sample data set traversal is stated to complete；
Step d：Count number that different sexes occurs in letter on each cis-position of user personality domain name and/or adjacent
The number that different sexes occurs in monogram on several cis-positions, and calculate alphabetical and/or adjacent several suitable on each cis-position
The probability that different sexes occurs in monogram on position.
In this step, according to expression formula
Calculate separately the probability that each letter on each cis-position corresponds to male with each monogram on adjacent several cis-positions.Its
In, the P (n-gram corresponds to male) on the left of equation is that length is that a monogram corresponds to male's on adjacent several cis-positions of n
Probability, P (n-gram corresponds to male) is the probability that a letter corresponds to male on single cis-position when n is 1；N- on the right side of equation
It is that a letter or length are that a monogram corresponds to man on adjacent several cis-positions of n on single cis-position that gram, which corresponds to male's frequency,
Property number, n-gram correspond to women frequency be on single cis-position one letter or length be on adjacent several cis-positions of n one letter
Combination corresponds to the number of women.
Step 104, using the probability as refer to parameter, the user personality domain name of unknown subscriber's gender is analyzed,
Judge user's gender；
This step specifically includes：
Step a：The length for obtaining the user personality domain name of unknown subscriber's gender, is denoted as k；
Step b：According to expression formula
The gender for calculating the user is the probability of male, wherein url indicates the part that user specifies in individual character domain name,
substr（Url, j, i）Indicate that jth position character starts the substring that the adjacent character that length is i is constituted in url, n is indicated
The number of substr (url, j, i), whIndicate the weight of the letter or monogram, (substr (url, j, i) is in sample number by P
Male is corresponded to according to concentrating) indicate that jth position character or jth position character start the sub- character that the adjacent character that length is i is constituted in url
The corresponding male's probability of letter or monogram on string；
Step c：The ratio for the male that result of calculation and step 102 in comparison step b are calculated；
Step d：When result of calculation in stepb is more than or equal to the ratio that step 102 is calculated, judgement is described unknown
The gender of gender user is male；
Step e：When result of calculation in stepb is less than the ratio that step 102 is calculated, the unknown gender is judged
The gender of user is women.
Below in conjunction with the accompanying drawings, the embodiment of the present invention two is illustrated.
An embodiment of the present invention provides a kind of user's gender analysis devices, and structure is as shown in Fig. 2, include：
Sampling module 201, is used for collecting sample data set, and the sample data set includes multipair user personality domain name and right
The user's gender answered；
Reference parameter computing module 202, for counting each cis-position in the user personality domain name that the sample data is concentrated
The probability that different monograms occur according to gender on different letters and adjacent several cis-positions；
Analysis module 203, ratio and the probability for concentrating male using the sample data are used as with reference to parameter, right
The user personality domain name of unknown subscriber's gender is analyzed, and judges user's gender.
Preferably, which further includes：
With reference to ratio computing module 204, the ratio of male is concentrated for calculating the sample data.
Preferably, the structure of the reference parameter computing module 202 is as shown in figure 3, include：
Gender extraction unit 2021 for taking the part that user specifies in a user personality domain name, while recording the use
The corresponding user's gender of family individual character domain name；
Counting unit 2022, if for letter occurs on each cis-position of the specified part number and/or adjacent
The number that different monograms occur on dry cis-position is counted；
Statistic unit 2023, whole user personality domain names for being concentrated to the sample data carry out the counting unit
Processing count on each cis-position of user personality domain name letter for dissimilarity until sample data set traversal is completed
The number that different sexes occurs in the monogram on number and/or adjacent several cis-positions not occurred, and calculate each cis-position
The probability that different sexes occurs in monogram on upper letter and/or adjacent several cis-positions.
Preferably, the statistic unit 2023 calculates on each cis-position monogram pair on letter and/or adjacent several cis-positions
It is specially in the probability that different sexes occur：
According to expression formula
Calculate separately each letter and the corresponding male's probability of each monogram on adjacent several cis-positions on each cis-position, wherein P
(n-gram corresponds to male) is that length is the probability that a monogram corresponds to male on adjacent several cis-positions of n, P when n is 1
(n-gram corresponds to male) is the probability that a letter corresponds to male on single cis-position, and it is single suitable that n-gram, which corresponds to male's frequency,
A monogram corresponds to the number of male on the upper letter in position or adjacent several cis-positions that length is n, and n-gram corresponds to women
Frequency is the number that a letter or length are a monogram corresponds to women on adjacent several cis-positions of n on single cis-position.
Preferably, the structure of the analysis module 203 is as shown in figure 4, include：
Domain name length acquiring unit 2031, the length of the user personality domain name for obtaining unknown subscriber's gender, note
For k；
Probability calculation unit 2032, for according to expression formula
The gender for calculating the user is the probability of male, wherein url indicates the part that user specifies in individual character domain name,
substr（Url, j, i）Indicate that jth position character starts the substring that the adjacent character that length is i is constituted in url, n is indicated
The number of substr (url, j, i), whIndicate the weight of the letter or monogram, (substr (url, j, i) is in sample number by P
Male is corresponded to according to concentrating) indicate that jth position character or jth position character start the sub- character that the adjacent character that length is i is constituted in url
The corresponding male's probability of letter or monogram on string；
Comparing unit 2033, the result of calculation for comparing probability calculation unit 2032 and reference ratio computing module 204
The ratio being calculated；
Judging unit 2034, for being to compare probability calculation unit 2032 in 2033 result of the comparison of the comparing unit
When result of calculation is more than or equal to the ratio being calculated with reference to ratio computing module 204, the property of the unknown gender user is judged
It Wei not male.
Preferably, the judging unit 2034 is additionally operable in 2033 result of the comparison of the comparing unit be to compare probability
When the result of calculation of computing unit is less than the ratio being calculated with reference to ratio computing module, the judgement unknown gender is used
The gender at family is women.
Below in conjunction with the accompanying drawings, the embodiment of the present invention three is illustrated.
The embodiment of the invention discloses a kind of user's gender analysis system, the individual character for being applied according to user, possessing or using
Domain name automatically classifies to the gender of user.The embodiment of the present invention passes through the means such as user data statistics, business associate first
The sample data set of the correspondence of individual character domain name and user's gender is obtained, the portion that user specifies in individual character domain name is then analyzed
Point, using the method for machine learning, train the grader classified to user's gender using individual character domain name.When needs are not to
When knowing that the individual character domain name of user's gender is classified, this grader is used, you can export user's gender of prediction.
It is as follows.
Step 1：The sample data set of the correspondence of individual character domain name and user's gender is acquired, analyzes and is used in individual character domain name
The specified part in family.
Step 2：Calculate the ratio shared by the sample data concentration male.
Step 3：The part that user specifies in property domain name one by one is taken, is denoted as character string one, while recording corresponding user's property
Not.
Step 4：The length of character string one is denoted as k, counts the frequency of occurrences of all 1-gram, 2- in character string one
The frequency of occurrences of gram, the frequency of occurrences of 3-gram until k-gram the frequency of occurrences（K represents the length of character string, Ke Yiwei
1 or 1 or more integer, the value upper limit of k）, by corresponding n-gram（N represents the length of character string, can be 1 to k）Appearance
It is cumulative that frequency presses corresponding user's gender.
Step 5：Step 3 is repeated, until the sample data set traversal acquired in step 1 is completed.
Step 6：Calculate the probability and the n-gram of user's gender corresponding to the n-gram occurred goes out occurrence
Number, while the probability that different sexes occur in statistical sample data set, collectively as the parameter of grader.
Step 6：When using grader, to the individual character domain name of unknown subscriber's gender, the part that wherein user specifies is analyzed,
Length is denoted as k, obtains its 1-gram until k-gram, the probability that its gender is male is calculated by following formula：
The gender for calculating the user is the probability of male, wherein url indicates the part that user specifies in individual character domain name,
substr（Url, j, i）Indicate that jth position character starts the substring that length is i in url, n indicates substr's (url, j, i)
Number, whIndicate the weight of the monogram.
Step 7：If the probability being calculated in step 6, which is more than the sample data being calculated in step 2, concentrates male
The individual character domain name can be then classified as corresponding male user, otherwise correspond to female user by the ratio of appearance.
Below in conjunction with the accompanying drawings, the embodiment of the present invention four is illustrated.
An embodiment of the present invention provides a kind of user's gender analysis method, detailed process is as follows：
Step 1：Collect following three individual character domain names：http://weibo.com/nickleave, http://
Weibo.com/inferpku, http:The part that //t.qq.com/bankofdota, wherein user are specified is respectively
nickleave、inferpku、bankofdota.In this example, system of the present invention by the means of business associate from
Information is obtained at the service provider of weibo.com and t.qq.com, learns that user's gender corresponding to nickleave is female,
User's gender corresponding to inferpku and bankofdota is man.
Step 2：Calculate the probability of male in all samples.In step 1, we have collected three samples, property in total
It Fen Biewei not nickleave（Female）、inferpku（Man）And bankofdota（Man）.It can be seen that in three samples, man
Sex ratio accounts for 2/3.
Step 3：Take nickleave, female.
Step 4：Nickleave corresponding 1-gram, 2-gram, 3-gram, 4-gram, 5-gram, 6-gram, 7-
Gram, 8-gram, 9-gram are added in corresponding women, and statistical result is as shown in table 1.In order to indicate convenient, only arranged in following table
Go out tri- kinds of situations of 1-gram, 2-gram, 3-gram.
Table 1
Step 5：Inferpku is repeated the above process.After cumulative, the result in table 1 is updated to table 2：
Table 2
| 1-gram | Male | Women | 2-gram | Male | Women | 3-gram | Male | Women |
| Frequency | Frequency | Frequency | Frequency | Frequency | Frequency | |||
| n | 1 | 1 | ni | 0 | 1 | nic | 0 | 1 |
| i | 1 | 1 | ic | 0 | 1 | ick | 0 | 1 |
| c | 0 | 1 | ck | 0 | 1 | ckl | 0 | 1 |
| k | 1 | 1 | kl | 0 | 1 | kle | 0 | 1 |
| l | 0 | 1 | le | 0 | 1 | lea | 0 | 1 |
| e | 1 | 2 | ea | 0 | 1 | eav | 0 | 1 |
| a | 0 | 1 | av | 0 | 1 | ave | 0 | 1 |
| v | 0 | 1 | ve | 0 | 1 | inf | 1 | 0 |
| f | 1 | 0 | in | 1 | 0 | nfe | 1 | 0 |
| r | 1 | 0 | nf | 1 | 0 | fer | 1 | 0 |
| p | 1 | 0 | fe | 1 | 0 | erp | 1 | 0 |
| u | 1 | 0 | er | 1 | 0 | rpk | 1 | 0 |
| rp | 1 | 0 | pku | 1 | 0 | |||
| pk | 1 | 0 | ||||||
| ku | 1 | 0 |
Bankofdota is repeated the above process again.After cumulative, table 2 is updated to table 3：
Table 3
Step 6：Calculate the probability and the n-gram of user's gender corresponding to the n-gram occurred goes out occurrence
Number.Wherein, n-gram corresponds to the probability that user's gender is male（Male's probability i.e. in following table）Computational methods be：
For example, it is to correspond to male's frequency 2 and women with 1-gram n in upper table that the 1-gram n in table 4, which correspond to male's probability,
What frequency 1 was calculated, i.e. 2/ (2+1)=0.666667.
In order to indicate convenient, tri- kinds of situations of 1-gram, 2-gram, 3-gram are only listed in table 4.
Table 4
| 1-gram | Male's probability | 2-gram | Male's probability | 3-gram | Male's probability |
| n | 0.666667 | ni | 0 | nic | 0 |
| i | 0.5 | ic | 0 | ick | 0 |
| c | 0 | ck | 0 | ckl | 0 |
| k | 0.666667 | kl | 0 | kle | 0 |
| l | 0 | le | 0 | lea | 0 |
| e | 0.333333 | ea | 0 | eav | 0 |
| a | 0.666667 | av | 0 | ave | 0 |
| v | 0 | ve | 0 | inf | 1 |
| f | 1 | in | 1 | nfe | 1 |
| r | 1 | nf | 1 | fer | 1 |
| p | 1 | fe | 1 | erp | 1 |
| u | 1 | er | 1 | rpk | 1 |
| b | 1 | rp | 1 | pku | 1 |
| o | 1 | pk | 1 | ban | 1 |
| d | 1 | ku | 1 | ank | 1 |
| t | 1 | ba | 1 | nko | 1 |
| an | 1 | kof | 1 | ||
| nk | 1 | ofd | 1 | ||
| ko | 1 | fdo | 1 | ||
| of | 1 | dot | 1 | ||
| fd | 1 | ota | 1 | ||
| do | 1 | ||||
| ot | 1 | ||||
| ta | 1 |
Step 7：Assuming that the individual character domain name classified is www.renren.com/eleven, wherein user is specified
Part be eleven, the n-gram occurred in eleven includes e（Three times）、l（Once）、v（Once）、n（Once）、le
（Once）、ve（Once）, and the n-gram occurred in above-mentioned third gender frequency meter lattice includes e（Three times）、l（Once）、v
（Once）、n （Three times）、le（Once）、ve（Once）.The n-gram that user name eleven occurs in gender frequency table is total
Number is 10.Thus the weight w of letter or monogram is calculatedh.According to above formula, the above numerical value is brought into, can obtain：
Step 7：
Step 8：Due to the eleven being calculated in upper step 2 correspond to user's gender be male probability 0.166 it is small
Male's proportion 0.67 in sample, therefore www.renren.com/eleven can be classified as to corresponding female user.
The embodiment provides a kind of user's gender analysis method and apparatus, collecting sample data set, the samples
Notebook data collection includes multipair user personality domain name and corresponding user's gender, counts the user personality domain that the sample data is concentrated
The different letters probability that different monograms occur according to gender on adjacent several cis-positions on each cis-position in name, then with described general
Rate, which is used as, refers to parameter, analyzes the user personality domain name of unknown subscriber's gender, judges user's gender, it realizes
User's gender analysis based on automation algorithm, it is more flexible and accurate, it solves existing analysis mode and is not suitable for individual character domain
First name and last name name is associated with the problem of weaker occasion.
Technical solution provided in an embodiment of the present invention is avoided by a kind of algorithm of automation to Praxeology number
According to the dependence in library.Dependence of the existing analysis mode to name, be not particularly suited for individual character domain name etc. be associated with name it is weaker
Occasion, and this problem is not present in technical solution provided in an embodiment of the present invention.In addition, the embodiment of the present invention passes through to individual character
The analysis of domain name can be used for showing in the broader practices such as advertisement optimization.
One of ordinary skill in the art will appreciate that all or part of step of above-described embodiment can use computer journey
Sequence flow realizes that the computer program can be stored in a computer readable storage medium, the computer program exists
On corresponding hardware platform（Such as system, unit, device）Execute, when being executed, include the steps that embodiment of the method it
One or combinations thereof.
Optionally, all or part of step of above-described embodiment can also realize that these steps can using integrated circuit
To be fabricated to integrated circuit modules one by one respectively, or by them multiple modules or step be fabricated to single integrated electricity
Road module is realized.In this way, the present invention is not limited to any specific hardware and softwares to combine.
General computing device may be used to realize in each device/function module/functional unit in above-described embodiment, it
Can be concentrated on a single computing device, can also be distributed on network constituted by multiple computing devices.
Each device/function module/functional unit in above-described embodiment realized in the form of software function module and as
Independent product sale in use, can be stored in a computer read/write memory medium.Computer mentioned above
Read/write memory medium can be read-only memory, disk or CD etc..
Any one skilled in the art in the technical scope disclosed by the present invention, can readily occur in variation or
It replaces, should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with the guarantor described in claim
It protects subject to range.
Claims (10)
1. a kind of user's gender analysis method, which is characterized in that including：
Collecting sample data set, the sample data set include multipair user personality domain name and corresponding user's gender；
Count different on different alphabetical and adjacent several cis-positions on each cis-position in the user personality domain name that the sample data is concentrated
The probability that monogram occurs according to gender, including：
Step a：The part that user specifies in a user personality domain name is taken, while recording the corresponding user of the user personality domain name
Gender；
Step b：To different letters on the number and/or adjacent several cis-positions of letter appearance on each cis-position of the specified part
The number that combination occurs is counted；
Step c：Processing such as step a to b is carried out to whole user personality domain names that the sample data is concentrated, until the sample
Notebook data collection traversal is completed；
Step d：Count number that different sexes occurs in letter on each cis-position of user personality domain name and/or adjacent several
The number that different sexes occurs in monogram on cis-position, and calculate on each cis-position on letter and/or adjacent several cis-positions
The probability that different sexes occurs in monogram；
Using the ratio of sample data concentration male and the probability as parameter is referred to, to the user of unknown subscriber's gender
Property domain name is analyzed, and judges user's gender.
2. user's gender analysis method according to claim 1, which is characterized in that the statistics sample data is concentrated
User personality domain name on each cis-position on different letters and adjacent several cis-positions different monograms occur according to gender it is general
Before the step of rate, further include：
Calculate the ratio that the sample data concentrates male.
3. user's gender analysis method according to claim 1, which is characterized in that it is each to count the user personality domain name
There are different sexes in the monogram on number and/or adjacent several cis-positions that different sexes occurs in letter on cis-position
Number, and calculate on each cis-position the probability tool that different sexes occurs in monogram on letter and/or adjacent several cis-positions
Body is：
According to expression formula
Calculate separately the probability that each letter on each cis-position corresponds to male with each monogram on adjacent several cis-positions；Wherein, etc.
P (n-gram corresponds to male) on the left of formula is that length is the probability that monogram on adjacent several cis-positions of n corresponds to male,
P (n-gram corresponds to male) corresponds to the probability of male for the letter on single cis-position when n is 1；N-gram pairs on the right side of equation
Answer that male's frequency is letter on single cis-position or length is time that monogram on adjacent several cis-positions of n corresponds to male
Number, n-gram correspond to that women frequency is letter on single cis-position or length is monogram pair on adjacent several cis-positions of n
It should be the number of women.
4. user's gender analysis method according to claim 1, which is characterized in that using the probability as refer to parameter,
The user personality domain name of unknown subscriber's gender is analyzed, judges that user's gender includes：
Step a：The length for obtaining the user personality domain name of unknown subscriber's gender, is denoted as k；
Step b：According to expression formula
The gender for calculating the user is the probability of male, wherein url indicates the part that user specifies in individual character domain name；
Substr (url, j, i) indicates that jth position character starts what the monogram on adjacent several cis-positions that length is i was constituted in url
Substring is the substring that the letter on single cis-position is constituted when i is 1；N indicates the number of substr (url, j, i)；wh
Indicate the weight of the letter or monogram；P (substr (url, j, i) concentrates corresponding male in sample data) indicates above-mentioned son
The corresponding male's probability of letter or monogram in character string；
Step c：Result of calculation and the sample data in comparison step b concentrate the ratio of male；
Step d：When result of calculation in stepb is more than or equal to the ratio that step c is calculated, the judgement unknown gender is used
The gender at family is male.
5. user's gender analysis method according to claim 4, which is characterized in that after the step d, further include：
Step e：When result of calculation in stepb is less than the ratio that step c is calculated, judge the unknown gender user's
Gender is women.
6. a kind of user's gender analysis device, which is characterized in that including：
Sampling module is used for collecting sample data set, and the sample data set includes multipair user personality domain name and corresponding use
Family gender；
Reference parameter computing module, for counting, difference is alphabetical on each cis-position in the user personality domain name that the sample data is concentrated
The probability that different monograms occur according to gender on adjacent several cis-positions；
Analysis module, the ratio and the probability for concentrating male using the sample data are as parameter is referred to, to unknown use
The user personality domain name of family gender is analyzed, and judges user's gender；
Wherein, the reference parameter computing module includes：
Gender extraction unit for taking the part that user specifies in a user personality domain name, while recording the user personality domain
The corresponding user's gender of name；
Counting unit, for the number and/or adjacent several cis-positions to letter appearance on each cis-position of the specified part
The number that different monograms occur is counted；
Statistic unit, the processing for whole user personality domain names that the sample data is concentrated to be carried out with the counting unit,
Until sample data set traversal is completed, count what different sexes occurred in letter on each cis-position of user personality domain name
The number that different sexes occurs in monogram on number and/or adjacent several cis-positions, and calculate on each cis-position letter and/
Or the probability that different sexes occurs in monogram on adjacent several cis-positions.
7. user's gender analysis device according to claim 6, which is characterized in that the device further includes：
With reference to ratio computing module, the ratio of male is concentrated for calculating the sample data.
8. user's gender analysis device according to claim 6, which is characterized in that the statistic unit calculates on each cis-position
The probability that different sexes occurs in monogram on alphabetical and/or adjacent several cis-positions is specially：
According to expression formula
Calculate separately each letter and the corresponding male's probability of each monogram on adjacent several cis-positions on each cis-position, wherein P (n-
Gram corresponds to male) be length it is the probability that a monogram corresponds to male on adjacent several cis-positions of n, P (n- when n is 1
Gram corresponds to male) it is the probability that a letter corresponds to male on single cis-position, it is single cis-position that n-gram, which corresponds to male's frequency,
A monogram corresponds to the number of male on a upper letter or adjacent several cis-positions that length is n, and n-gram corresponds to women frequency
Rate is the number that a letter or length are a monogram corresponds to women on adjacent several cis-positions of n on single cis-position.
9. user's gender analysis device according to claim 7, which is characterized in that the analysis module includes：
Domain name length acquiring unit, the length of the user personality domain name for obtaining unknown subscriber's gender, is denoted as k；
Probability calculation unit, for according to expression formula
The gender for calculating the user is the probability of male, wherein url indicates the part that user specifies in individual character domain name,
Substr (url, j, i) indicates that jth position character starts the substring that the adjacent character that length is i is constituted in url, and n is indicated
The number of substr (url, j, i), whIndicate the weight of the letter or monogram, (substr (url, j, i) is in sample number by P
Male is corresponded to according to concentrating) indicate that jth position character or jth position character start the sub- character that the adjacent character that length is i is constituted in url
The corresponding male's probability of letter or monogram on string；
Comparing unit, for the result of calculation for comparing probability calculation unit and the ratio being calculated with reference to ratio computing module；
Judging unit, for being to compare the result of calculation of probability calculation unit to be more than or equal in the comparing unit result of the comparison
When the ratio being calculated with reference to ratio computing module, judge that the gender of the unknown gender user is male.
10. user's gender analysis device according to claim 9, which is characterized in that
The judging unit, be additionally operable to the comparing unit result of the comparison be compare probability calculation unit result of calculation it is small
When the ratio being calculated with reference to ratio computing module, judge that the gender of the unknown gender user is women.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310526980.4A CN104598452B (en) | 2013-10-30 | 2013-10-30 | User's gender analysis method and apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310526980.4A CN104598452B (en) | 2013-10-30 | 2013-10-30 | User's gender analysis method and apparatus |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104598452A CN104598452A (en) | 2015-05-06 |
| CN104598452B true CN104598452B (en) | 2018-09-11 |
Family
ID=53124251
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310526980.4A Active CN104598452B (en) | 2013-10-30 | 2013-10-30 | User's gender analysis method and apparatus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104598452B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106656943B (en) * | 2015-11-03 | 2019-09-17 | 秒针信息技术有限公司 | A kind of matching process and device of network user's attribute |
| CN105809557A (en) * | 2016-03-15 | 2016-07-27 | 微梦创科网络科技（中国）有限公司 | Method and device for mining genders of users in social network |
| CN106844687B (en) * | 2017-01-23 | 2021-01-01 | 炫彩互动网络科技有限公司 | Method and system for determining gender of user based on game log |
| CN107357782B (en) * | 2017-06-29 | 2020-12-18 | 深圳市金立通信设备有限公司 | Method and terminal for identifying gender of user |
| CN111309913A (en) * | 2020-02-26 | 2020-06-19 | 北京慧博科技有限公司 | Method for analyzing gender by name |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1987852A (en) * | 2005-12-21 | 2007-06-27 | 腾讯科技（深圳）有限公司 | Method and device for determining communication object attribute according to news content |
| CN103164470A (en) * | 2011-12-15 | 2013-06-19 | 盛大计算机（上海）有限公司 | Directional application method based on user gender distinguished results and system thereof |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8005782B2 (en) * | 2007-08-10 | 2011-08-23 | Microsoft Corporation | Domain name statistical classification using character-based N-grams |
| US8041662B2 (en) * | 2007-08-10 | 2011-10-18 | Microsoft Corporation | Domain name geometrical classification using character-based n-grams |
-
2013
- 2013-10-30 CN CN201310526980.4A patent/CN104598452B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1987852A (en) * | 2005-12-21 | 2007-06-27 | 腾讯科技（深圳）有限公司 | Method and device for determining communication object attribute according to news content |
| CN103164470A (en) * | 2011-12-15 | 2013-06-19 | 盛大计算机（上海）有限公司 | Directional application method based on user gender distinguished results and system thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104598452A (en) | 2015-05-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6568609B2 (en) | Grammar model for structured search queries | |
| CN102929939B (en) | The offer method and device of customized information | |
| CN104598452B (en) | User's gender analysis method and apparatus | |
| US20150081431A1 (en) | Posterior probability calculating apparatus, posterior probability calculating method, and non-transitory computer-readable recording medium | |
| US20120066195A1 (en) | Search assist powered by session analysis | |
| WO2021160157A1 (en) | Group display method and device | |
| CN107292463A (en) | A kind of method and system that the project evaluation is carried out to application program | |
| CN107291755B (en) | Terminal pushing method and device | |
| CN110472154A (en) | A kind of resource supplying method, apparatus, electronic equipment and readable storage medium storing program for executing | |
| CN104915426B (en) | Information sorting method, the method and device for generating information sorting model | |
| CN101000611A (en) | Method for providing and inquiry information for public by interconnection network | |
| CN106650760A (en) | Method and device for recognizing user behavioral object based on flow analysis | |
| US11574123B2 (en) | Content analysis utilizing general knowledge base | |
| CN110275952A (en) | News recommended method, device and medium based on user's short-term interest | |
| CN107690634A (en) | Automatic query pattern generation | |
| CN106326338A (en) | Service providing method and device based on search engine | |
| CN110019837B (en) | User portrait generation method and device, computer equipment and readable medium | |
| CN112182391A (en) | User portrait drawing method and device | |
| CN106910135A (en) | User recommends method and device | |
| CN111782816B (en) | Method and device for generating knowledge graph, searching method, engine and system | |
| US20170124120A1 (en) | Information processing system, information processing method, and information processing program | |
| US8560468B1 (en) | Learning expected values for facts | |
| CN109144999B (en) | Data positioning method, device, storage medium and program product | |
| EP4198758B1 (en) | Method and system for scalable acceleration of data processing pipeline | |
| CN108595453B (en) | URL (Uniform resource locator) identifier mapping obtaining method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20150506 Assignee: Beijing Interactive Technology Co., Ltd. Assignor: Beijing Sibotu Information Technology Co., Ltd. Contract record no.: 2015110000019 Denomination of invention: Method and device for analyzing user gender License type: Exclusive License Record date: 20150603 |
|
| LICC | Enforcement, change and cancellation of record of contracts on the licence for exploitation of a patent or utility model | ||
| EC01 | Cancellation of recordation of patent licensing contract |
Assignee: Beijing Interactive Technology Co., Ltd. Assignor: The second hand information technology Co. Ltd. Contract record no.: 2015110000019 Date of cancellation: 20160426 |
|
| EM01 | Change of recordation of patent licensing contract |
Change date: 20160426 Contract record no.: 2015110000019 Assignor after: The second hand information technology Co. Ltd. Assignor before: Beijing Sibotu Information Technology Co., Ltd. |
|
| LICC | Enforcement, change and cancellation of record of contracts on the licence for exploitation of a patent or utility model | ||
| CB02 | Change of applicant information |
Address after: 100102 Beijing, Chaoyang District Fu Tong East Street, building 1, room 5, room 321008 Applicant after: The second hand information technology Co. Ltd. Address before: Beijing City, a small town east of Changping District road 102218 in No. 398 Coal Construction Group No. 1 building, 4 floor second hand system Applicant before: Beijing Sibotu Information Technology Co., Ltd. |
|
| COR | Change of bibliographic data | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |