CN104717674A

CN104717674A - Number attribute recognition method and device, terminal and server

Info

Publication number: CN104717674A
Application number: CN201410721351.1A
Authority: CN
Inventors: 周楠; 左平地; 常富洋; 谢冉; 秦吉胜; 李振博
Original assignee: Beijing Qihoo Technology Co Ltd; Qizhi Software Beijing Co Ltd
Current assignee: Beijing Qihoo Technology Co Ltd; Qizhi Software Beijing Co Ltd
Priority date: 2014-12-02
Filing date: 2014-12-02
Publication date: 2015-06-17

Abstract

The invention discloses a number attribute recognition method and device, a terminal and a server, and mainly relates to the technical field of communications. The main purpose is to recognize the corresponding attributes of numbers. The method includes the steps that the behavior data corresponding to the first number to be recognized are acquired, and the behavior data corresponding to the first number are the data generated in the communication behavioral process of a user whom the first number belongs to; the behavior data corresponding to the first number are calculated so as to acquire the attribute of the first number through the behavior data corresponding to the known second number and a recognition model trained by the corresponding attribute; the behavior data corresponding to the second number are the data generated in the communication behavioral process of a user whom the second number belongs to. According to the number attribute recognition method and device, the terminal and the server, the purposes of the users whom the numbers belong to can be reflected accurately through the number attributes determined by the behavior data corresponding to the numbers, and thus the numbers can be correspondingly processed.

Description

Number attribute recognition methods, device, terminal and server

Technical field

The present invention relates to communication technical field, in particular to a kind of number attribute recognition methods, device, terminal and server.

Background technology

At present, the popularization degree of the terminal equipments such as mobile phone is more and more higher, and each terminal user has unique number, the communication of can and carry out between other people conversing, send by this number user various ways such as short breath, mail.

The terminal equipments such as mobile phone come easily simultaneously for user, also bring many problems: increase from the note of unknown number and phone, it comprises refuse messages and harassing call, for user brings puzzlement.The mode of current identification unknown number is, user, after the note receiving unknown number and phone, according to the content of phone or note, marks unknown number, such as, is labeled as " swindle ", " distribution " etc.; Collect this flag data and preserve; Then when other users also receive note or the phone of this unknown number, point out this unknown number for " swindle " or " distribution " etc. according to the flag data preserved to other users.

The defect of such scheme is: the marking behavior too relying on user, and it is very few or mark inaccurate possibility that the marking behavior of user also exists mark number of times, be then difficult to accurately identify unknown number according to the flag data collected.

Summary of the invention

In view of the above problems, the present invention is proposed to provide a kind of overcoming the problems referred to above or the number attribute recognition methods solved the problem at least in part, device, terminal and server.

According to one aspect of the present invention, provide a kind of number attribute recognition methods, it comprises: obtain the behavioral data corresponding to the first number to be identified, the data produced in the communication behavior process that the user of behavioral data belonging to described first number that described first number is corresponding carries out; Use the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, the behavioral data corresponding to described first number calculates, to obtain the attribute of described first number; The data produced in the communication behavior process that the user of behavioral data belonging to described second number that wherein said second number is corresponding carries out.

According to another aspect of the present invention, provide a kind of number attribute recognition device, it comprises: behavioral data acquisition module, for obtaining the behavioral data corresponding to the first number to be identified, the data produced in the communication behavior process that the user of behavioral data belonging to described first number that described first number is corresponding carries out; Attribute Recognition module, for using the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, the behavioral data corresponding to described first number calculates, to obtain the attribute of described first number; The data produced in the communication behavior process that the user of behavioral data belonging to described second number that wherein said second number is corresponding carries out.

According to another aspect of the invention, provide a kind of terminal, it comprises: aforesaid number attribute recognition device, for the behavioral data according to described number, identifies the attribute of described number.

According to an also aspect of the present invention, provide a kind of server, it comprises: behavioral data receiver module, the behavioral data that the number to be identified for receiving self terminal is corresponding; Aforesaid number attribute recognition device, for the behavioral data according to described number, identifies the attribute of described number; Attribute sending module, sends to described terminal by the attribute of described number.

According to above technical scheme, known number attribute recognition methods of the present invention, device, terminal and server at least will have the following advantages:

Number owning user carries out in the process of communication behavior based on different objects, and the data produced are necessarily different, so behavioral data corresponding to number to reflect that user carries out the object of communication behavior; So according to the attribute of the determined number of the behavioral data that number is corresponding, the object of number owning user accurately can be embodied, so that carry out respective handling to this number.

Above-mentioned explanation is only the general introduction of technical solution of the present invention, in order to technological means of the present invention can be better understood, and can be implemented according to the content of specification, and can become apparent, below especially exemplified by the specific embodiment of the present invention to allow above and other objects of the present invention, feature and advantage.

Accompanying drawing explanation

By reading hereafter detailed description of the preferred embodiment, various other advantage and benefit will become cheer and bright for those of ordinary skill in the art.Accompanying drawing only for illustrating the object of preferred implementation, and does not think limitation of the present invention.And in whole accompanying drawing, represent identical parts by identical reference symbol.In the accompanying drawings:

Fig. 1 shows the flow chart of number attribute recognition methods according to an embodiment of the invention;

Figure 1A shows the operating diagram of number attribute recognition methods according to an embodiment of the invention;

Fig. 2 shows the flow chart of number attribute recognition methods according to an embodiment of the invention;

Fig. 3 shows the flow chart of number attribute recognition methods according to an embodiment of the invention;

Fig. 4 shows the flow chart of number attribute recognition methods according to an embodiment of the invention;

Fig. 5 shows the flow chart of number attribute recognition methods according to an embodiment of the invention;

Fig. 6 shows the flow chart of number attribute recognition methods according to an embodiment of the invention;

Fig. 7 shows the flow chart of number attribute recognition methods according to an embodiment of the invention;

Fig. 8 shows the flow chart of number attribute recognition methods according to an embodiment of the invention;

Fig. 9 shows the flow chart of number attribute recognition methods according to an embodiment of the invention;

Figure 10 shows the flow chart of number attribute recognition methods according to an embodiment of the invention;

Figure 11 shows the flow chart of number attribute recognition methods according to an embodiment of the invention;

Figure 12 shows the block diagram of number attribute recognition device according to an embodiment of the invention;

Figure 13 shows the block diagram of number attribute recognition device according to an embodiment of the invention;

Figure 14 shows the block diagram of number attribute recognition device according to an embodiment of the invention;

Figure 15 shows the block diagram of number attribute recognition device according to an embodiment of the invention;

Figure 16 shows the block diagram of number attribute recognition device according to an embodiment of the invention;

Figure 17 shows the block diagram of number attribute recognition device according to an embodiment of the invention;

Figure 18 shows the block diagram of terminal according to an embodiment of the invention;

Figure 19 shows the block diagram of server according to an embodiment of the invention.

Embodiment

Below with reference to accompanying drawings exemplary embodiment of the present disclosure is described in more detail.Although show exemplary embodiment of the present disclosure in accompanying drawing, however should be appreciated that can realize the disclosure in a variety of manners and not should limit by the embodiment set forth here.On the contrary, provide these embodiments to be in order to more thoroughly the disclosure can be understood, and complete for the scope of the present disclosure can be conveyed to those skilled in the art.

As shown in Figure 1, disclose a kind of number attribute recognition methods in one embodiment of the present of invention, it comprises:

Step 110, obtains the behavioral data corresponding to the first number to be identified, the data produced in the communication behavior process that the user of behavioral data belonging to the first number that the first number is corresponding carries out.In the present embodiment, the type of communication behavior do not limited, include but not limited to make a phone call, send short messages, send out mail etc.; Behavioral data do not limited, includes but not limited to air time, the duration of call, be hung up accounting, incoming call accounting, address book contact accounting, non-toll message accounting, request call time to the average request number of times of the other side.

Step 120, uses the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, and the behavioral data corresponding to the first number calculates, to obtain the attribute of the first number; The data produced in the communication behavior process that the user of the behavioral data that wherein the second number is corresponding belonging to the second number carries out.In the present embodiment, do not limit attribute, it includes but not limited to be the label or classification etc. that the first number increases.In the present embodiment, the algorithm corresponding to model of cognition does not limit, and such as, svm, boosting decision tree scheduling algorithm is all applicable.According to the technical scheme of the present embodiment, number owning user carries out in the process of communication behavior based on different objects, and the data produced are necessarily different, so behavioral data corresponding to number to reflect that user carries out the object of communication behavior; So according to the attribute of the determined number of the behavioral data that number is corresponding, the object of number owning user accurately can be embodied, so that carry out respective handling to this number.

Such as, according to Fig. 1, collect the incoming call accounting (behavioral data) of the call behavior (communication behavior) of 200 numbers (the second number), and the label that each number is corresponding (attribute); Model of cognition is generated based on boosting decision Tree algorithms, after getting the incoming call accounting (behavioral data) of the call behavior (communication behavior) of number A (the first number), inputted model of cognition, and from model of cognition, export label corresponding to number A (attribute), such as, model of cognition is very low based on the incoming call accounting of number A, so can determine it is express delivery food delivery phone, add " express delivery food delivery " label, this label may be used for showing in the terminals such as mobile phone, such as shown in Figure 1A, illustrate that this incoming call is express delivery food delivery phone.

As shown in Figure 2, disclose a kind of number attribute recognition methods in one embodiment of the present of invention, it comprises:

Step 210, obtains the text message of the first number owning user transmission, and from text message, extracts word as behavioral data corresponding to the first number.In the present embodiment, do not limit the type of text message, it includes but not limited to note, mail etc.The present embodiment does not limit the mode extracting word, can only extract fixing word, extract after also can using the segmenter participle of prior art again.

Step 220, uses the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, and the behavioral data corresponding to the first number calculates, to obtain the attribute of the first number; The data produced in the communication behavior process that the user of the behavioral data that wherein the second number is corresponding belonging to the second number carries out.According to the technical scheme of the present embodiment, text message comprises a lot of crucial word, can embody the object that the first number sends text message, so be conducive to according to the word in text message the attribute determining the first number.

Such as, according to Fig. 2, the user through statistics number B have issued a note (text message); Rule of thumb the content of known a lot of harassing and wrecking note is all relevant with " drawing a bill ", so can arrange the mode extracting word is extraction " invoice " this word, the number with " invoice " wording is black number; Successfully extract " invoice " (word) from this note after, under this number B being categorized into " black number " classification (attribute) according to " invoice ".

As shown in Figure 3, disclose a kind of number attribute recognition methods in one embodiment of the present of invention, it comprises:

Step 310, according to the filtering characters storehouse of presetting, filters the character in text message; The character needing to filter is have recorded in filtering characters storehouse.In the present embodiment, to needing the character filtered out not limit, can be polytype punctuation mark.

Step 320, extracts word as behavioral data corresponding to the first number from text message.In the present embodiment, do not limit the type of text message, it includes but not limited to note, mail etc.The present embodiment does not limit the mode extracting word, can only extract fixing word, extract after also can using the segmenter participle of prior art again.

Step 330, uses the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, and the behavioral data corresponding to the first number calculates, to obtain the attribute of the first number; The data produced in the communication behavior process that the user of the behavioral data that wherein the second number is corresponding belonging to the second number carries out.According to the technical scheme of the present embodiment, a lot of rubbish short message is identified for avoiding, and usually adds the character much playing interference effect in the information, after therefore filtering these characters, easilier can extract key words from text message.

Such as, according to Fig. 3, in the note (text message) that number C sends, include following content: " ... exploitation | ticket ... "; In the filtering characters storehouse of presetting, regulation need by | filtered symbol falls, then after filtering the content of note be " ... draw a bill ... "; Carry out participle extraction to the note after filtering, can extract smoothly word " invoice " (word), this note belongs to harassing and wrecking note, thus number C can be categorized under " harassing and wrecking number ".

As shown in Figure 4, disclose a kind of number attribute recognition methods in one embodiment of the present of invention, it comprises:

Step 410, the character conversion storehouse according to presetting inquires about whether have character to be converted in text message, and changes according to character conversion storehouse when finding character to be converted; Character to be converted is recorded in character conversion storehouse, and the character after the conversion corresponding with character to be converted.In the present embodiment, character conversion storehouse may be used for Chinese-character digital to be converted to Arabic numerals.

Step 420, extracts word as behavioral data corresponding to the first number from text message.In the present embodiment, do not limit the type of text message, it includes but not limited to note, mail etc.The present embodiment does not limit the mode extracting word, can only extract fixing word, extract after also can using the segmenter participle of prior art again.

Step 430, uses the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, and the behavioral data corresponding to the first number calculates, to obtain the attribute of the first number; The data produced in the communication behavior process that the user of the behavioral data that wherein the second number is corresponding belonging to the second number carries out.According to the technical scheme of the present embodiment, a lot of rubbish short message is identified for avoiding, and the character conversion some easily identified is other forms, is changed, easilier can extract key words from text message by the technical scheme of the present embodiment to it.

Such as, according to Fig. 4, in the mail (text message) that number D sends, include following content: " ... 6587324 ... "; This mail is sent to cloud server from terminal after being encrypted by rivest, shamir, adelman, the character conversion storehouse that cloud server is preset, have recorded need conversion Chinese-character digital, and conversion after Arabic numerals, then conversion after mail be " ... 6587324 ... "; Participle extraction is carried out to the mail after conversion, " 6587324 " (word) can be extracted smoothly, the server in high in the clouds adopts model of cognition to identify, can find that this number is for swindle number, so number D can be categorized under " swindle number ".

As shown in Figure 5, disclose a kind of number attribute recognition methods in one embodiment of the present of invention, it comprises:

Step 510, the information obtaining the behavior of repeatedly conversing that the first number owning user carries out is as behavioral data corresponding to the first number.The behavioral data that first number is corresponding comprises following at least one: air time of behavior of repeatedly conversing, the duration of call, when being hung up accounting, incoming call accounting, address book contact accounting, non-toll message accounting, request call to the average request number of times of the other side.

Step 520, uses the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, and the behavioral data corresponding to the first number calculates, to obtain the attribute of the first number; The data produced in the communication behavior process that the user of the behavioral data that wherein the second number is corresponding belonging to the second number carries out.According to the technical scheme of the present embodiment, in call behavior, much can reflect that this call is normal talking or occurs the situation such as swindle, distribution, so be conducive to according to the information of the behavior of call the attribute determining the first number consumingly.

Such as, according to Fig. 5, the user of number E has dialed No. 20 phones (call behavior) in one day, the number of times be hung up is 16 times, be hung up after number of times is encrypted by rivest, shamir, adelman and be sent to cloud server, cloud server analysis is known is hung up accounting 80% (information); According to the model of cognition of cloud server training, when being hung up accounting more than 60%, judge that number can shield; So number E can be added in " blacklist " (attribute) classification.

As shown in Figure 6, disclose a kind of number attribute recognition methods in one embodiment of the present of invention, it comprises:

Step 610, obtains the behavioral data corresponding to the first number to be identified, the data produced in the communication behavior process that the user of behavioral data belonging to the first number that the first number is corresponding carries out.In the present embodiment, the type of communication behavior do not limited, it includes but not limited to make a phone call, send short messages, send out mail etc.; Behavioral data do not limited, includes but not limited to air time, the duration of call, be hung up accounting, incoming call accounting, address book contact accounting, non-toll message accounting, request call time to the average request number of times of the other side.

Step 620, obtains the attribute of the second number, for training identification module.In the present embodiment, do not limit the attribute of the second number, it includes but not limited to be the label or classification etc. that the second number increases.

Step 630, uses the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, and the behavioral data corresponding to the first number calculates, to obtain the attribute of the first number; The data produced in the communication behavior process that the user of the behavioral data that wherein the second number is corresponding belonging to the second number carries out.In the present embodiment, do not limit attribute, it includes but not limited to be the label or classification etc. that the first number increases.In the present embodiment, the algorithm corresponding to model of cognition does not limit, and such as, svm, boosting decision tree scheduling algorithm is all applicable.

As shown in Figure 7, disclose a kind of number attribute recognition methods in one embodiment of the present of invention, it comprises:

Step 710, obtains the behavioral data corresponding to the first number to be identified, the data produced in the communication behavior process that the user of behavioral data belonging to the first number that the first number is corresponding carries out.

Step 720, obtain the flag data that the second number of producing in special time period is corresponding, flag data is for representing that the second number is labeled as the first attribute by other users.In the present embodiment, do not limit the type of the first attribute, it can be any attribute allowing user's mark.

Step 730, according to flag data, calculates the second number in special time period and is marked as the number of times of the first attribute.In the present embodiment, do not limit special time period, it can be the time period of any length.

Step 740, according to the size of number of times, determines whether the second number has the first attribute.In the present embodiment, the number of times be labeled in certain hour section is too much, can determine that this second number has the first attribute.

Step 750, uses the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, and the behavioral data corresponding to the first number calculates, to obtain the attribute of the first number; The data produced in the communication behavior process that the user of the behavioral data that wherein the second number is corresponding belonging to the second number carries out.According to the technical scheme of the present embodiment, utilize user to the mark of the second number, determine the attribute of the second number, order of accuarcy is very high.

Such as, according to Fig. 7, learning that to choose (special time period) internal labeling in nearest two weeks be the number of times of " food delivery " (attribute) number for number F through statistics is 45 times; Internal labeling in two weeks is that the number of times of same attribute is greater than and then represents that it is more active for 30 times, and the mark of user has reference value, so determine that the attribute of number F is " food delivery " (attribute) number, may be used for training.

As shown in Figure 8, disclose a kind of number attribute recognition methods in one embodiment of the present of invention, it comprises:

Step 810, obtains the behavioral data corresponding to the first number to be identified, the data produced in the communication behavior process that the user of behavioral data belonging to the first number that the first number is corresponding carries out.

Step 820, has been identified as having the second attribute at the second number, and when cannot get flag data corresponding to the second number of producing in special time period, determines that the second number has the second attribute; Flag data for represent the second number be labeled as by other users the first attribute.In the present embodiment, do not limit the first attribute and the second attribute, it includes but not limited to be the label or classification etc. that the second number increases.

Step 830, uses the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, and the behavioral data corresponding to the first number calculates, to obtain the attribute of the first number; The data produced in the communication behavior process that the user of the behavioral data that wherein the second number is corresponding belonging to the second number carries out.According to the technical scheme of the present embodiment, be identified as the second attribute at the second number, come to negate this recognition result as long as no user's mark, namely can determine that this second number has the second attribute, accuracy rate is very high.

Such as, according to Fig. 8, for the number G being identified as " personal number " classification (attribute), if nearest two weeks (special time period) does not have user that number G is labeled as other class numbers, such as, " food delivery ", " harassing and wrecking " class number; Then can determine that current recognition result is accurately, so determine that " personal number " classification of number G may be used for training.

As shown in Figure 9, disclose a kind of number attribute recognition methods in one embodiment of the present of invention, it comprises:

Step 910, obtains the behavioral data corresponding to the first number to be identified, the data produced in the communication behavior process that the user of behavioral data belonging to the first number that the first number is corresponding carries out.

Step 920, when model of cognition is multiple, adds up the quantity of each attribute in the analysis result of multiple model of cognition, selects the attribute of the first number according to quantity height from analysis result.According to the technical scheme of the present embodiment, the recognition result of comprehensive multiple model of cognition, is conducive to the attribute accurately selecting the first number.

Such as, according to Fig. 9, the recognition result obtained is identified by the four kinds of model of cognition the being positioned at cloud server H that checks numbers---" personal number ", " personal number ", " personal number " and " harassing and wrecking " (attribute), according to recognition result design ballot: " personal number " 3 ticket, " harassing and wrecking " 1 ticket, so determine that number H belongs to " personal number " classification.

As shown in Figure 10, disclose a kind of number attribute recognition methods in one embodiment of the present of invention, it comprises:

Step 1010, obtains the behavioral data corresponding to the first number to be identified, the data produced in the communication behavior process that the user of behavioral data belonging to the first number that the first number is corresponding carries out.

Step 1020, the often kind behavioral data corresponding according to the first number, calculates the probability that the first number has the first attribute.In the present embodiment, do not limit the type of the first attribute, it includes but not limited to be the label or classification etc. that the first number increases.

Step 1030, according to each self-corresponding probability of multiple behavioral datas that the first number is corresponding, calculates common corresponding joint probability.In the present embodiment, under joint probability reflects the simultaneous situation of multiple behavior, whether the first number is the probability of the first attribute.

Step 1040, according to the size of joint probability, judges whether the first number has the first attribute.By the technical scheme of the present embodiment, the possibility of the first attribute can be had by reasonable assessment first number, not think that when possibility is lower first number has the first attribute, to prevent misjudgment.

Such as, according to Figure 10, through statistics in the recent repeatedly call behavior (communication behavior) of number I, in conversation object, address book contact accounting is 20% (behavioral data), and incoming call accounting is 30% (behavioral data); According to address list contact accounting, judge number I as the probability of " harassing and wrecking class " (first attribute) number be 60%, according to incoming call accounting judge number I as the probability of " harassing and wrecking class " number be 50%; Calculating joint probability is 80%; Preset joint probability higher than 90% time, can determine that corresponding number is " harassing and wrecking class " number, so now can not determine that number I is " harassing and wrecking class " number.

As shown in figure 11, disclose a kind of number attribute recognition methods in one embodiment of the present of invention, it comprises:

Step 1110, by the time period of presetting, obtains the behavioral data that the first number of producing within each time period is corresponding.In the present embodiment, do not limit the length of time period, such as, a time period can be one hour or one day.

Step 1120, by the cycle length comprising multiple time period preset, the behavioral data that the first number produced in the corresponding time period in cumulative multiple cycle is corresponding.In the present embodiment, do not limit the length in cycle, such as, one-period can be one week or one day.

Step 1130, according to each cumulative behavioral data obtained, identifies the attribute of the first number.In the present embodiment, if same communication behavior occurs in the diverse time period, may produce diverse behavioral data, the behavioral data therefore produced in the diverse time period accumulates together, and can not embody the feature of user within each time period on the contrary; But there is the larger problem of contingency in the behavioral data obtained in the single time period; And the behavioral data by producing in cumulative corresponding multiple time periods in the present embodiment, the feature of the plurality of time period can be embodied, overcome again the problem of contingency.

Such as, according to Figure 11, in one week, statistics is with hour (time period) the exhalation number of times (behavioral data) for unit statistics number J, and during 11 .-12 points (corresponding time period) in cumulative every day (cycle) exhalation number of times---15 times, 20 times, 20 times, 15 times, 20 times, 10 times, 10 times, the cumulative exhalation number of times obtaining one week interior 11 .-12 number J is 110 times; Because 11 .-12 these time periods are the food delivery time, so preset rules is designed to 11 .-12 exhalation number of times in a week, more than 80 times, can judge that respective number is " food delivery " (attribute) number; So known number J belongs to " food delivery " class number.

As shown in figure 12, disclose a kind of number attribute recognition device in one embodiment of the present of invention, it comprises:

Behavioral data acquisition module 1210, obtains the behavioral data corresponding to the first number to be identified, the data produced in the communication behavior process that the user of behavioral data belonging to the first number that the first number is corresponding carries out.In the present embodiment, the type of communication behavior do not limited, include but not limited to make a phone call, send short messages, send out mail etc.; Behavioral data do not limited, includes but not limited to air time, the duration of call, be hung up accounting, incoming call accounting, address book contact accounting, non-toll message accounting, request call time to the average request number of times of the other side.

Attribute Recognition module 1220, uses the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, and the behavioral data corresponding to the first number calculates, to obtain the attribute of the first number; The data produced in the communication behavior process that the user of the behavioral data that wherein the second number is corresponding belonging to the second number carries out.In the present embodiment, do not limit attribute, it includes but not limited to be the label or classification etc. that the first number increases.In the present embodiment, the algorithm corresponding to model of cognition does not limit, and such as, svm, boosting decision tree scheduling algorithm is all applicable.According to the technical scheme of the present embodiment, number owning user carries out in the process of communication behavior based on different objects, and the data produced are necessarily different, so behavioral data corresponding to number to reflect that user carries out the object of communication behavior; So according to the attribute of the determined number of the behavioral data that number is corresponding, the object of number owning user accurately can be embodied, so that carry out respective handling to this number.

Such as, according to Figure 12, collect the incoming call accounting (behavioral data) of the call behavior (communication behavior) of 200 numbers (the second number), and the label that each number is corresponding (attribute); Model of cognition is generated based on boosting decision Tree algorithms, after getting the incoming call accounting (behavioral data) of the call behavior (communication behavior) of number A (the first number), inputted model of cognition, and from model of cognition, export label corresponding to number A (attribute), such as, model of cognition is very low based on the incoming call accounting of number A, so can determine it is express delivery food delivery phone, add " express delivery food delivery " label, this label may be used for showing in the terminals such as mobile phone, such as shown in Figure 1A, illustrate that this incoming call is express delivery food delivery phone.

Disclose a kind of number attribute recognition device in one embodiment of the present of invention, it comprises:

Behavioral data acquisition module 1210, obtains the text message of the first number owning user transmission, and from text message, extracts word as behavioral data corresponding to the first number.In the present embodiment, do not limit the type of text message, it includes but not limited to note, mail etc.The present embodiment does not limit the mode extracting word, can only extract fixing word, extract after also can using the segmenter participle of prior art again.

Attribute Recognition module 1220, uses the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, and the behavioral data corresponding to the first number calculates, to obtain the attribute of the first number; The data produced in the communication behavior process that the user of the behavioral data that wherein the second number is corresponding belonging to the second number carries out.According to the technical scheme of the present embodiment, text message comprises a lot of crucial word, can embody the object that the first number sends text message, so be conducive to according to the word in text message the attribute determining the first number.

Such as, according to Figure 12, the user through statistics number B have issued a note (text message); Rule of thumb the content of known a lot of harassing and wrecking note is all relevant with " drawing a bill ", so can arrange the mode extracting word is extraction " invoice " this word, the number with " invoice " wording is black number; Successfully extract " invoice " (word) from this note after, under this number B being categorized into " black number " classification (attribute) according to " invoice ".

As shown in figure 13, disclose a kind of number attribute recognition device in one embodiment of the present of invention, it comprises:

Filtering module 1310, according to the filtering characters storehouse of presetting, filters the character in text message; The character needing to filter is have recorded in filtering characters storehouse.In the present embodiment, to needing the character filtered out not limit, can be polytype punctuation mark.

Behavioral data acquisition module 1320, extracts word as behavioral data corresponding to the first number from text message.In the present embodiment, do not limit the type of text message, it includes but not limited to note, mail etc.The present embodiment does not limit the mode extracting word, can only extract fixing word, extract after also can using the segmenter participle of prior art again.

Attribute Recognition module 1330, uses the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, and the behavioral data corresponding to the first number calculates, to obtain the attribute of the first number; The data produced in the communication behavior process that the user of the behavioral data that wherein the second number is corresponding belonging to the second number carries out.According to the technical scheme of the present embodiment, a lot of rubbish short message is identified for avoiding, and usually adds the character much playing interference effect in the information, after therefore filtering these characters, easilier can extract key words from text message.

Such as, according to Figure 13, in the note (text message) that number C sends, include following content: " ... exploitation | ticket ... "; In the filtering characters storehouse of presetting, regulation need by | filtered symbol falls, then after filtering the content of note be " ... draw a bill ... "; Carry out participle extraction to the note after filtering, can extract smoothly word " invoice " (word), this note belongs to harassing and wrecking note, thus number C can be categorized under " harassing and wrecking number ".

As shown in figure 14, disclose a kind of number attribute recognition device in one embodiment of the present of invention, it comprises:

Modular converter 1410, the character conversion storehouse according to presetting inquires about whether have character to be converted in text message, and changes according to character conversion storehouse when finding character to be converted; Character to be converted is recorded in character conversion storehouse, and the character after the conversion corresponding with character to be converted.In the present embodiment, character conversion storehouse may be used for Chinese-character digital to be converted to Arabic numerals.

Behavioral data acquisition module 1420, extracts word as behavioral data corresponding to the first number from text message.In the present embodiment, do not limit the type of text message, it includes but not limited to note, mail etc.The present embodiment does not limit the mode extracting word, can only extract fixing word, extract after also can using the segmenter participle of prior art again.

Attribute Recognition module 1430, uses the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, and the behavioral data corresponding to the first number calculates, to obtain the attribute of the first number; The data produced in the communication behavior process that the user of the behavioral data that wherein the second number is corresponding belonging to the second number carries out.According to the technical scheme of the present embodiment, a lot of rubbish short message is identified for avoiding, and the character conversion some easily identified is other forms, is changed, easilier can extract key words from text message by the technical scheme of the present embodiment to it.

Such as, according to Figure 14, in the mail (text message) that number D sends, include following content: " ... 6587324 ... "; This mail is sent to cloud server from terminal after being encrypted by rivest, shamir, adelman, the character conversion storehouse that cloud server is preset, have recorded need conversion Chinese-character digital, and conversion after Arabic numerals, then conversion after mail be " ... 6587324 ... "; Participle extraction is carried out to the mail after conversion, " 6587324 " (word) can be extracted smoothly, the server in high in the clouds adopts model of cognition to identify, can find that this number is for swindle number, so number D can be categorized under " swindle number ".

Behavioral data acquisition module 1210, the information obtaining the behavior of repeatedly conversing that the first number owning user carries out is as behavioral data corresponding to the first number.The behavioral data that first number is corresponding comprises following at least one: air time of behavior of repeatedly conversing, the duration of call, when being hung up accounting, incoming call accounting, address book contact accounting, non-toll message accounting, request call to the average request number of times of the other side.

Attribute Recognition module 1220, uses the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, and the behavioral data corresponding to the first number calculates, to obtain the attribute of the first number; The data produced in the communication behavior process that the user of the behavioral data that wherein the second number is corresponding belonging to the second number carries out.According to the technical scheme of the present embodiment, in call behavior, much can reflect that this call is normal talking or occurs the situation such as swindle, distribution, so be conducive to according to the information of the behavior of call the attribute determining the first number consumingly.

Such as, according to Figure 12, the user of number E has dialed No. 20 phones (call behavior) in one day, the number of times be hung up is 16 times, be hung up after number of times is encrypted by rivest, shamir, adelman and be sent to cloud server, cloud server analysis is known is hung up accounting 80% (information); According to the model of cognition of cloud server training, when being hung up accounting more than 60%, judge that number can shield; So number E can be added in " blacklist " (attribute) classification.

As shown in figure 15, disclose a kind of number attribute recognition device in one embodiment of the present of invention, it comprises:

Behavioral data acquisition module 1510, obtains the behavioral data corresponding to the first number to be identified, the data produced in the communication behavior process that the user of behavioral data belonging to the first number that the first number is corresponding carries out.In the present embodiment, the type of communication behavior do not limited, it includes but not limited to make a phone call, send short messages, send out mail etc.; Behavioral data do not limited, includes but not limited to air time, the duration of call, be hung up accounting, incoming call accounting, address book contact accounting, non-toll message accounting, request call time to the average request number of times of the other side.

Attribute acquisition module 1520, obtains the attribute of the second number, for training identification module.In the present embodiment, do not limit the attribute of the second number, it includes but not limited to be the label or classification etc. that the second number increases.

Attribute Recognition module 1530, uses the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, and the behavioral data corresponding to the first number calculates, to obtain the attribute of the first number; The data produced in the communication behavior process that the user of the behavioral data that wherein the second number is corresponding belonging to the second number carries out.In the present embodiment, do not limit attribute, it includes but not limited to be the label or classification etc. that the first number increases.In the present embodiment, the algorithm corresponding to model of cognition does not limit, and such as, svm, boosting decision tree scheduling algorithm is all applicable.

As shown in figure 16, disclose a kind of number attribute recognition device in one embodiment of the present of invention, it comprises:

Behavioral data acquisition module 1610, obtains the behavioral data corresponding to the first number to be identified, the data produced in the communication behavior process that the user of behavioral data belonging to the first number that the first number is corresponding carries out.

Flag data acquisition module 1620, obtain the flag data that the second number of producing in special time period is corresponding, flag data is for representing that the second number is labeled as the first attribute by other users.In the present embodiment, do not limit the type of the first attribute, it can be any attribute allowing user's mark.

Mark number of times computing module 1630, according to flag data, calculates the second number in special time period and is marked as the number of times of the first attribute.In the present embodiment, do not limit special time period, it can be the time period of any length.

Attribute acquisition module 1640, according to the size of number of times, determines whether the second number has the first attribute.In the present embodiment, the number of times be labeled in certain hour section is too much, can determine that this second number has the first attribute.

Attribute Recognition module 1650, uses the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, and the behavioral data corresponding to the first number calculates, to obtain the attribute of the first number; The data produced in the communication behavior process that the user of the behavioral data that wherein the second number is corresponding belonging to the second number carries out.According to the technical scheme of the present embodiment, utilize user to the mark of the second number, determine the attribute of the second number, order of accuarcy is very high.

Such as, according to Figure 16, learning that to choose (special time period) internal labeling in nearest two weeks be the number of times of " food delivery " (attribute) number for number F through statistics is 45 times; Internal labeling in two weeks is that the number of times of same attribute is greater than and then represents that it is more active for 30 times, and the mark of user has reference value, so determine that the attribute of number F is " food delivery " (attribute) number, may be used for training.

Behavioral data acquisition module 1510, obtains the behavioral data corresponding to the first number to be identified, the data produced in the communication behavior process that the user of behavioral data belonging to the first number that the first number is corresponding carries out.

Attribute acquisition module 1520, has been identified as having the second attribute at the second number, and when cannot get flag data corresponding to the second number of producing in special time period, determines that the second number has the second attribute; Flag data for represent the second number be labeled as by other users the first attribute.In the present embodiment, do not limit the first attribute and the second attribute, it includes but not limited to be the label or classification etc. that the second number increases.

Attribute Recognition module 1530, uses the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, and the behavioral data corresponding to the first number calculates, to obtain the attribute of the first number; The data produced in the communication behavior process that the user of the behavioral data that wherein the second number is corresponding belonging to the second number carries out.According to the technical scheme of the present embodiment, be identified as the second attribute at the second number, come to negate this recognition result as long as no user's mark, namely can determine that this second number has the second attribute, accuracy rate is very high.

Such as, according to Figure 15, for the number G being identified as " personal number " classification (attribute), if nearest two weeks (special time period) does not have user that number G is labeled as other class numbers, such as, " food delivery ", " harassing and wrecking " class number; Then can determine that current recognition result is accurately, so determine that " personal number " classification of number G may be used for training.

Behavioral data acquisition module 1210, obtains the behavioral data corresponding to the first number to be identified, the data produced in the communication behavior process that the user of behavioral data belonging to the first number that the first number is corresponding carries out.

Attribute Recognition module 1220, when model of cognition is multiple, adds up the quantity of each attribute in the analysis result of multiple model of cognition, selects the attribute of the first number according to quantity height from analysis result.According to the technical scheme of the present embodiment, the recognition result of comprehensive multiple model of cognition, is conducive to the attribute accurately selecting the first number.

Such as, according to Figure 12, the recognition result obtained is identified by the four kinds of model of cognition the being positioned at cloud server H that checks numbers---" personal number ", " personal number ", " personal number " and " harassing and wrecking " (attribute), according to recognition result design ballot: " personal number " 3 ticket, " harassing and wrecking " 1 ticket, so determine that number H belongs to " personal number " classification.

As shown in figure 17, disclose a kind of number attribute recognition device in one embodiment of the present of invention, it comprises:

Behavioral data acquisition module 1710, obtains the behavioral data corresponding to the first number to be identified, the data produced in the communication behavior process that the user of behavioral data belonging to the first number that the first number is corresponding carries out.

First probability evaluation entity 1720, the often kind behavioral data corresponding according to the first number, calculates the probability that the first number has the first attribute.In the present embodiment, do not limit the type of the first attribute, it includes but not limited to be the label or classification etc. that the first number increases.

Second probability evaluation entity 1730, according to each self-corresponding probability of multiple behavioral datas that the first number is corresponding, calculates common corresponding joint probability.In the present embodiment, under joint probability reflects the simultaneous situation of multiple behavior, whether the first number is the probability of the first attribute.

Attribute Recognition module 1740, according to the size of joint probability, judges whether the first number has the first attribute.By the technical scheme of the present embodiment, the possibility of the first attribute can be had by reasonable assessment first number, not think that when possibility is lower first number has the first attribute, to prevent misjudgment.

Such as, according to Figure 17, through statistics in the recent repeatedly call behavior (communication behavior) of number I, in conversation object, address book contact accounting is 20% (behavioral data), and incoming call accounting is 30% (behavioral data); According to address list contact accounting, judge number I as the probability of " harassing and wrecking class " (first attribute) number be 60%, according to incoming call accounting judge number I as the probability of " harassing and wrecking class " number be 50%; Calculating joint probability is 80%; Preset joint probability higher than 90% time, can determine that corresponding number is " harassing and wrecking class " number, so now can not determine that number I is " harassing and wrecking class " number.

Behavioral data acquisition module 1210, by the time period of presetting, obtains the behavioral data that the first number of producing within each time period is corresponding.In the present embodiment, do not limit the length of time period, such as, a time period can be one hour or one day.And by the cycle length comprising multiple time period preset, the behavioral data that the first number produced in the corresponding time period in cumulative multiple cycle is corresponding.In the present embodiment, do not limit the length in cycle, such as, one-period can be one week or one day.

Attribute Recognition module 1220, according to each cumulative behavioral data obtained, identifies the attribute of the first number.In the present embodiment, if same communication behavior occurs in the diverse time period, may produce diverse behavioral data, the behavioral data therefore produced in the diverse time period accumulates together, and can not embody the feature of user within each time period on the contrary; But there is the larger problem of contingency in the behavioral data obtained in the single time period; And the behavioral data by producing in cumulative corresponding multiple time periods in the present embodiment, the feature of the plurality of time period can be embodied, overcome again the problem of contingency.

Such as, according to Figure 12, in one week, statistics is with hour (time period) the exhalation number of times (behavioral data) for unit statistics number J, and during 11 .-12 points (corresponding time period) in cumulative every day (cycle) exhalation number of times---15 times, 20 times, 20 times, 15 times, 20 times, 10 times, 10 times, the cumulative exhalation number of times obtaining one week interior 11 .-12 number J is 110 times; Because 11 .-12 these time periods are the food delivery time, so preset rules is designed to 11 .-12 exhalation number of times in a week, more than 80 times, can judge that respective number is " food delivery " (attribute) number; So known number J belongs to " food delivery " class number.

As shown in figure 18, provide a kind of terminal in one embodiment of the present of invention, it comprises: the number attribute recognition device in any embodiment that Figure 12 to Figure 17 is corresponding.Terminal in the present embodiment includes but not limited to mobile phone, flat computer etc., the number attribute recognition device provided by previous embodiment, when terminal receives incoming call or the information of unknown number, can identify the attribute of unknown number in time; This attribute can be shown to user with reference to whether receiving the incoming call of unknown number or read the information of unknown number; Also can realize, to location number incoming call or the automatic process of information, such as, being identified as black number then automatically by its incoming call blocking according to this attribute, and by its information screen.

As shown in figure 19, provide a kind of server in one embodiment of the present of invention, it comprises: behavioral data receiver module 1910, the behavioral data that the number to be identified for receiving self terminal is corresponding.Number attribute recognition device in any embodiment that Figure 12 to Figure 17 is corresponding, for the behavioral data according to number, identifies the attribute of number.Attribute sending module 1920, sends to terminal by the attribute of number, to show the attribute of number in terminal.

According to the technical scheme of the present embodiment, based on the number attribute recognition device that previous embodiment provides, when terminal receives incoming call or the information of unknown number, unknown number can be sent to server, identified the attribute of unknown number by server in time, and inform terminal; This attribute can be shown to user with reference to whether receiving the incoming call of unknown number or read the information of unknown number; Also can realize, to location number incoming call or the automatic process of information, such as, being identified as black number then automatically by its incoming call blocking according to this attribute, and by its information screen.

Intrinsic not relevant to any certain computer, virtual system or miscellaneous equipment with display at this algorithm provided.Various general-purpose system also can with use based on together with this teaching.According to description above, the structure constructed required by this type systematic is apparent.In addition, the present invention is not also for any certain programmed language.It should be understood that and various programming language can be utilized to realize content of the present invention described here, and the description done language-specific is above to disclose preferred forms of the present invention.

In specification provided herein, describe a large amount of detail.But can understand, embodiments of the invention can be put into practice when not having these details.In some instances, be not shown specifically known method, structure and technology, so that not fuzzy understanding of this description.

Similarly, be to be understood that, in order to simplify the disclosure and to help to understand in each inventive aspect one or more, in the description above to exemplary embodiment of the present invention, each feature of the present invention is grouped together in single embodiment, figure or the description to it sometimes.But, the method for the disclosure should be construed to the following intention of reflection: namely the present invention for required protection requires feature more more than the feature clearly recorded in each claim.Or rather, as claims below reflect, all features of disclosed single embodiment before inventive aspect is to be less than.Therefore, the claims following embodiment are incorporated to this embodiment thus clearly, and wherein each claim itself is as independent embodiment of the present invention.

Those skilled in the art are appreciated that and adaptively can change the module in the equipment in embodiment and they are arranged in one or more equipment different from this embodiment.Module in embodiment or unit or assembly can be combined into a module or unit or assembly, and multiple submodule or subelement or sub-component can be put them in addition.Except at least some in such feature and/or process or unit be mutually repel except, any combination can be adopted to combine all processes of all features disclosed in this specification (comprising adjoint claim, summary and accompanying drawing) and so disclosed any method or equipment or unit.Unless expressly stated otherwise, each feature disclosed in this specification (comprising adjoint claim, summary and accompanying drawing) can by providing identical, alternative features that is equivalent or similar object replaces.

In addition, those skilled in the art can understand, although embodiments more described herein to comprise in other embodiment some included feature instead of further feature, the combination of the feature of different embodiment means and to be within scope of the present invention and to form different embodiments.Such as, in the following claims, the one of any of embodiment required for protection can use with arbitrary compound mode.

All parts embodiment of the present invention with hardware implementing, or can realize with the software module run on one or more processor, or realizes with their combination.It will be understood by those of skill in the art that the some or all functions that microprocessor or digital signal processor (DSP) can be used in practice to realize according to the some or all parts in the number attribute recognition device of the embodiment of the present invention.The present invention can also be embodied as part or all equipment for performing method as described herein or device program (such as, computer program and computer program).Realizing program of the present invention and can store on a computer-readable medium like this, or the form of one or more signal can be had.Such signal can be downloaded from internet website and obtain, or provides on carrier signal, or provides with any other form.

The present invention will be described instead of limit the invention to it should be noted above-described embodiment, and those skilled in the art can design alternative embodiment when not departing from the scope of claims.In the claims, any reference symbol between bracket should be configured to limitations on claims.Word " comprises " not to be got rid of existence and does not arrange element in the claims or step.Word "a" or "an" before being positioned at element is not got rid of and be there is multiple such element.The present invention can by means of including the hardware of some different elements and realizing by means of the computer of suitably programming.In the unit claim listing some devices, several in these devices can be carry out imbody by same hardware branch.Word first, second and third-class use do not represent any order.Can be title by these word explanations.

A1, a kind of number attribute recognition methods, it comprises:

Obtain the behavioral data corresponding to the first number to be identified, the data produced in the communication behavior process that the user of behavioral data belonging to described first number that described first number is corresponding carries out;

Use the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, the behavioral data corresponding to described first number calculates, to obtain the attribute of described first number; The data produced in the communication behavior process that the user of behavioral data belonging to described second number that wherein said second number is corresponding carries out.

A2, method according to A1, wherein, obtain the behavioral data corresponding to the first number to be identified, specifically comprise:

Obtain the text message of described first number owning user transmission, and from described text message, extract word as behavioral data corresponding to described first number.

A3, method according to A2, wherein, extract from described text message word as the behavioral data that described first number is corresponding before, also comprise:

According to the filtering characters storehouse of presetting, the character in described text message is filtered; The character needing to filter is have recorded in described filtering characters storehouse.

A4, method according to A2, wherein, extract from described text message word as the behavioral data that described first number is corresponding before, also comprise:

Character conversion storehouse according to presetting inquires about whether have character to be converted in described text message, and changes according to described character conversion storehouse when finding described character to be converted; Described character to be converted is recorded in described character conversion storehouse, and the character after the conversion corresponding with described character to be converted.

A5, method according to A1, wherein, obtain the behavioral data corresponding to the first number to be identified, specifically comprise:

The information obtaining the behavior of repeatedly conversing that described first number owning user carries out is as behavioral data corresponding to described first number.

A6, method according to A5, wherein, the behavioral data that described first number is corresponding comprises following at least one:

Air time of described behavior of repeatedly conversing, the duration of call, when being hung up accounting, incoming call accounting, address book contact accounting, non-toll message accounting, request call to the average request number of times of the other side.

A7, method according to A1, wherein, using the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, the behavioral data corresponding to described first number calculates, before the attribute obtaining described first number, also comprise:

Obtain the attribute of described second number, for the described identification module of training.

A8, method according to A7, wherein, obtain the attribute of described second number, specifically comprise:

The flag data that described second number that acquisition produces in special time period is corresponding, described flag data is for representing that described second number is labeled as the first attribute by other users;

According to described flag data, calculate described second number in described special time period and be marked as the number of times of described first attribute;

According to the size of described number of times, determine whether described second number has described first attribute.

A9, method according to A7, wherein, obtain the attribute of described second number, specifically comprise:

Be identified as that there is the second attribute at described second number, and when cannot get flag data corresponding to described second number that produces in special time period, determined that described second number has described second attribute; Described flag data for represent described second number be labeled as by other users the first attribute.

A10, method according to A1, wherein, described model of cognition is multiple; Use the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, the behavioral data corresponding to described first number calculates, and to obtain the attribute of described first number, specifically comprises:

Add up the quantity of each attribute in the analysis result of multiple described model of cognition, from described analysis result, select the attribute of described first number according to quantity height.

A11, method according to A1, wherein, the behavioral data that described first number is corresponding has multiple; Use the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, the behavioral data corresponding to described first number calculates, and to obtain the attribute of described first number, specifically comprises:

The often kind behavioral data corresponding according to described first number, calculates the probability that described first number has the first attribute;

According to each self-corresponding probability of multiple behavioral datas that described first number is corresponding, calculate common corresponding joint probability;

According to the size of described joint probability, judge whether described first number has described first attribute.

A12, method according to any one of A1 to A11, wherein, obtain the behavioral data corresponding to the first number to be identified, specifically comprise:

By the time period of presetting, obtain the behavioral data that described first number that produces within each time period is corresponding;

By the cycle length comprising multiple time period preset, the behavioral data that described first number produced in the corresponding time period in cumulative multiple cycle is corresponding;

Use the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, the behavioral data corresponding to described first number calculates, and to obtain the attribute of described first number, specifically comprises:

According to each cumulative behavioral data obtained, identify the attribute of described first number.

A13, a kind of number attribute recognition device, it comprises:

Behavioral data acquisition module, for obtaining the behavioral data corresponding to the first number to be identified, the data produced in the communication behavior process that the user of behavioral data belonging to described first number that described first number is corresponding carries out;

Attribute Recognition module, for using the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, the behavioral data corresponding to described first number calculates, to obtain the attribute of described first number; The data produced in the communication behavior process that the user of behavioral data belonging to described second number that wherein said second number is corresponding carries out.

A14, device according to A13, wherein,

Described behavioral data acquisition module obtains the text message of described first number owning user transmission, and from described text message, extracts word as behavioral data corresponding to described first number.

A15, device according to A14, wherein, also comprise:

Filtering module, for according to the filtering characters storehouse of presetting, filters the character in described text message; The character needing to filter is have recorded in described filtering characters storehouse.

A16, device according to A14, wherein, also comprise:

Modular converter, for inquiring about whether have character to be converted according to the character conversion storehouse of presetting in described text message, and changes according to described character conversion storehouse when finding described character to be converted; Described character to be converted is recorded in described character conversion storehouse, and the character after the conversion corresponding with described character to be converted.

A17, device according to A13, wherein, the information that described behavioral data acquisition module obtains the behavior of repeatedly conversing that described first number owning user carries out is as behavioral data corresponding to described first number.

A18, device according to A17, wherein, the behavioral data that described first number is corresponding comprises following at least one:

A19, device according to A13, wherein, also comprise:

Attribute acquisition module, for obtaining the attribute of described second number, for the described identification module of training.

A20, device according to A19, wherein, also comprise:

Flag data acquisition module, for obtaining flag data corresponding to described second number that produces in special time period, described flag data is for representing that described second number is labeled as the first attribute by other users;

Mark number of times computing module, for according to described flag data, calculates described second number in described special time period and is marked as the number of times of described first attribute;

Described attribute acquisition module, according to the size of described number of times, determines whether described second number has described first attribute.

A21, device according to A19, wherein,

Described Attribute Recognition module has been identified as having the second attribute at described second number, and when cannot get flag data corresponding to described second number that produces in special time period, determines that described second number has described second attribute; Described flag data for represent described second number be labeled as by other users the first attribute.

A22, device according to A13, wherein, described model of cognition is multiple;

In the analysis result of the multiple described model of cognition of described Attribute Recognition module statistics, the quantity of each attribute, selects the attribute of described first number from described analysis result according to quantity height.

A23, device according to A13, wherein, the behavioral data that described first number is corresponding has multiple; Also comprise:

First probability evaluation entity, for the often kind behavioral data corresponding according to described first number, calculates the probability that described first number has the first attribute;

Second probability evaluation entity, for each self-corresponding probability of the multiple behavioral datas corresponding according to described first number, calculates common corresponding joint probability;

Described Attribute Recognition module is used for the size according to described joint probability, judges whether described first number has described first attribute.

A24, device according to any one of A13 to A23, wherein, described behavioral data acquisition module, by the time period of presetting, obtains the behavioral data that described first number that produces within each time period is corresponding; And by the cycle length comprising multiple time period preset, the behavioral data that described first number produced in the corresponding time period in cumulative multiple cycle is corresponding;

Described Attribute Recognition module, according to each cumulative behavioral data obtained, identifies the attribute of described first number.

A25, a kind of terminal, it comprises:

Number attribute recognition device according to any one of A13 to A24.

A26, a kind of server, it comprises:

Behavioral data receiver module, the behavioral data that the number to be identified for receiving self terminal is corresponding;

Number attribute recognition device according to any one of A13 to A24, for the behavioral data according to described number, identifies the attribute of described number;

Attribute sending module, sends to described terminal by the attribute of described number.

Claims

1. a number attribute recognition methods, it comprises:

2. method according to claim 1, wherein, using the model of cognition of being trained by behavioral data corresponding to the second known number and attribute, the behavioral data corresponding to described first number calculates, before the attribute obtaining described first number, also comprise:

3. method according to claim 2, wherein, obtains the attribute of described second number, specifically comprises:

4. method according to claim 2, wherein, obtains the attribute of described second number, specifically comprises:

5. a number attribute recognition device, it comprises:

6. device according to claim 5, wherein, also comprises:

7. device according to claim 6, wherein, also comprises:

8. device according to claim 6, wherein,

9. a terminal, it comprises:

Number attribute recognition device according to any one of claim 5 to 8.

10. a server, it comprises:

Number attribute recognition device according to any one of claim 5 to 8, for the behavioral data according to described number, identifies the attribute of described number;