CN103744887A

CN103744887A - Method and device for people search and computer equipment

Info

Publication number: CN103744887A
Application number: CN201310717845.8A
Authority: CN
Inventors: 张超; 赵世奇
Original assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Current assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Priority date: 2013-12-23
Filing date: 2013-12-23
Publication date: 2014-04-23
Anticipated expiration: 2033-12-23
Also published as: CN103744887B

Abstract

The invention provides a method for people search. The method comprises the steps of obtaining a search sequence from a user; obtaining at least one search segmentation word from the search sequence; obtaining at least one candidate person based on the at least one search segmentation word, wherein each of all or part of the at least one search segmentation word can be matched with one or more people tags of one candidate person; sequencing the all or part of the candidate people in the at least one candidate person according to weight information people tags matched with the search segmentation words of all candidate people; providing the sequenced candidate people to the user.

Description

A kind of method, device and computer equipment for personage's search

Technical field

The present invention relates to computer network field, relate in particular to a kind of method, device and computer equipment for personage's search.

Background technology

In prior art, when user need to search plain personage, they are many by universal search engine, and input is used for searching for personage's keyword, and in Search Results, searches it and wish the information obtaining.But, because the universal search engine of prior art lacks the processing for personage's search, therefore the search engine of prior art cannot provide good personage's Search Results, as it tends to provide a large amount of, user may unwanted or inaccurate information.For example, that search engine tends to provide is a large amount of, not for personage's Search Results, makes user need in magnanimity result, inquire about the people information that it needs; Again for example, search engine may provide wrong people information, because, often owing to personage introduction and personage not being inconsistent such as factors such as mischief, malice mark, blow one's own trumpet, can brag about it as an ordinary people is that the famous cardiologist in the whole nation searched engine offer user in internet; Again for example, some professional personages may be only movable in minority website or community, makes it be difficult to searched engine and excavate and offer user etc.

Summary of the invention

The object of this invention is to provide a kind of for searching for personage's method, device and computer equipment.

According to an aspect of the present invention, provide a kind of for searching for personage's method, wherein, the method comprises the following steps:

A obtains the search sequence from user;

B is by obtaining at least one search participle in described search sequence;

C based on described at least one search participle obtain at least one candidate personage, wherein, described at least one search participle all or part of in each search participle all can with one or more coupling the in a candidate personage's people tag;

D, according to the weight information of people tag each candidate personage, that mate with described search participle, sorts to all or part of candidate personage in described at least one candidate personage;

E offers described user by the candidate personage after sequence.

According to another aspect of the present invention, also provide a kind of for searching for personage's searcher, wherein, this searcher comprises:

The first acquisition device, for obtaining the search sequence from user;

The second acquisition device, for being obtained at least one search participle by described search sequence;

The first determining device, for obtaining at least one candidate personage based on described at least one search participle, wherein, described at least one search participle all or part of in each search participle all can with one or more coupling the in a candidate personage's people tag;

Collator, for according to the weight information of people tag each candidate personage, that mate with described search participle, sorts to all or part of candidate personage in described at least one candidate personage;

Generator, for offering described user by the candidate personage after sequence.

Compared with prior art, the present invention has the following advantages: 1) can retrieve for personage, and the weight of people tag that can be based on mating with search sequence, candidate personage is sorted, thereby make the fiduciary level of personage's result for retrieval higher; 2) can determine the weight information of its people tag in conjunction with personage's community network activity or personal information, thereby further guarantee search based on people tag and the accuracy of sequence; 3) personage's search can be in universal search engine, realized, thereby personage's function of search of across a network community can be realized; 4) content of issuing by other personages associated with figure picture and personage self is upgraded label for this personage, make the reaction personage's that personage's label can be more effective, more complete feature, and the operations such as interpolation, deletion and the change by people tag make people tag can react more accurately the feature that personage is current had, reduced in personage's search and occurred that personage omits or the inaccurate situation of people tag, thereby the Search Results that personage is searched for is more comprehensive, also more meet user's search condition.

Accompanying drawing explanation

By reading the detailed description that non-limiting example is done of doing with reference to the following drawings, it is more obvious that other features, objects and advantages of the present invention will become:

Fig. 1 be one embodiment of the invention for searching for personage's method flow schematic diagram;

Fig. 2 be another embodiment of the present invention for searching for personage's method flow schematic diagram;

Fig. 3 be another embodiment of the present invention for searching for personage's method flow schematic diagram;

Fig. 4 be one embodiment of the invention for searching for personage's the structural representation of searcher;

Fig. 5 be another embodiment of the present invention for searching for personage's the structural representation of searcher;

Fig. 6 be another embodiment of the present invention for searching for personage's the structural representation of searcher.

In accompanying drawing, same or analogous Reference numeral represents same or analogous parts.

Embodiment

Below in conjunction with accompanying drawing, the present invention is described in further detail.

Fig. 1 be one embodiment of the invention for searching for personage's method flow schematic diagram.Wherein, the method for the present embodiment mainly realizes by computer equipment; Described computer equipment comprises the network equipment and subscriber equipment; The described network equipment includes but not limited to server group or the cloud consisting of a large amount of computing machines or the webserver based on cloud computing (Cloud Computing) of single network server, multiple webserver composition, wherein, cloud computing is the one of Distributed Calculation, the super virtual machine being comprised of the loosely-coupled computing machine collection of a group; Described subscriber equipment can include but not limited to PC, panel computer, smart mobile phone, PDA, IPTV etc.; The residing network of described computer equipment includes but not limited to internet, wide area network, Metropolitan Area Network (MAN), LAN (Local Area Network), VPN network etc.

It should be noted that, the described network equipment, subscriber equipment and network are only for giving an example, and other computing equipments existing or that may occur from now on or network, as applicable to the present invention, also should be included in protection domain of the present invention, and are contained in this with way of reference.

According to the method for the present embodiment, comprise step S1, step S2, step S3, step S4 and step S5.Preferably, according to the method for the present embodiment, can in universal search engine, realize, this universal search engine can obtain the people information of across a network community.

In step S1, computer equipment obtains the search sequence from user.

Wherein, computer equipment obtains from the mode of user's search sequence and includes but not limited to:

1) computer equipment is the subscriber equipment of direct and user interactions, and computer equipment directly obtains the search sequence of user's input.

For example, computer equipment is PC, and PC directly obtains the search sequence that user directly inputs in its browser searches input frame.

2) computer equipment obtains the search sequence of user's input by other computer equipments.

For example, computer equipment is the network equipment, and user is inputted search sequence in the browser searches input frame of subscriber equipment, and subscriber equipment offers the network equipment by this search sequence by network, and the network equipment is by this search sequence of Network Capture.

It should be noted that, above-mentioned for example only for technical scheme of the present invention is described better, but not limitation of the present invention it should be appreciated by those skilled in the art that any implementation of obtaining from user's search sequence, all should be within the scope of the present invention.

In step S2, computer equipment is by obtaining at least one search participle in described search sequence.

Particularly, computer equipment can adopt various ways, obtains at least one search participle from search sequence.

As a kind of optimal way as step S2, step S2 further comprises step S201 and step S202.

In step S201, computer equipment carries out participle to described search sequence, obtains participle list.

For example, computer equipment obtains the search sequence " lawyer of economic class " from user in step S1, computer equipment is according to semantic analysis and predetermined participle dictionary, this search sequence is carried out to participle, obtain participle list comprise " economic class ", " " and the participle list of " lawyer ".

In step S202, computer equipment is selected important participle from described participle list, as described at least one search participle.Wherein, computer equipment can determine whether the participle in participle list is important participle in several ways; For example computer equipment can be by mating the participle in participle list with the participle in predetermined participle dictionary, and/or, semantic analysis, the semantic analysis to the participle in participle list itself and the syntactic position analysis of the participle based in participle list in search sequence etc. based on to search sequence, determine whether the participle in participle list is important participle.

For example, in step S201, computer equipment is by search sequence " lawyer of economic class " is carried out to participle, obtains participle list, the participle wherein comprising be " economic class ", " " and " lawyer "; In step S202, computer equipment mates the participle in participle list with the participle in predetermined participle dictionary, and select important participle " economic class " and " lawyer " according to matching result, as corresponding two the search participles of search sequence " lawyer of economic class ".

It should be noted that, above-mentioned for example only for technical scheme of the present invention is described better, but not limitation of the present invention, those skilled in the art should understand that, any described search sequence is carried out to participle, obtain participle list, and select important participle from described participle list, as the implementation of described at least one search participle, all should be within the scope of the present invention.

And for example, as the another kind of optimal way of step S2, step S2 comprises step S2 '.

In step S2 ', computer equipment carries out participle to described search sequence, obtains described at least one search participle.

Wherein, the operation of the participle that computer equipment carries out search sequence, consider following at least one factor:

1) semantic difference of the different word segmentation result of described search sequence.

Preferably, computer equipment is selected the less participle mode of semantic ambiguity.If there is the subphrase larger with the semantic differential of this phrase in multiple subphrases that a phrase in search sequence obtains after being further divided, select this complete phrase as a participle, and do not select the subphrase of this phrase as participle.When and for example in search sequence, multiple phrases that do not comprise mutually can produce multiple combination, after Selection and Constitute with the less array mode of search sequence semantic difference etc.It should be noted that, search sequence can be directly a phrase.

For example, search sequence comprises phrase " i Phone ", multiple subphrases that this phrase obtains after being further divided are " apple " and " mobile phone ", computer equipment is by the pre-definite phrase classification of inquiry, the primary categories of determining phrase " i Phone " is " mobile phone ", and the primary categories of determining the subphrase " apple " of this phrase comprises " fruit " and " mobile phone ", the primary categories of subphrase " mobile phone " is " mobile phone ", in multiple subphrases that computer equipment judgement " i Phone " obtains after being further divided, there is the subphrase " apple " larger with the semantic differential of this phrase, computer equipment selects phrase " i Phone " as a participle, and do not select the subphrase " apple " of this phrase and " mobile phone " as participle.

2) matching result of described search sequence and predetermined participle dictionary.

Preferably, when the phrase that carries out participle acquisition when the granularity with less can not obtain the vocabulary matching with it in predetermined participle dictionary, the preferential phrase of selecting to carry out with larger granularity participle gained.If a phrase in search sequence, further having the subphrase that does not belong to predetermined participle dictionary in the rear multiple subphrases that obtain of segmentation, is directly selected this phrase, and is not selected the subphrase of this phrase, as participle.

For example, a phrase in search sequence is " Warcraft Howard ", obtains subphrase " Warcraft " and " Howard " after this phrase participle, but in predetermined participle dictionary, do not there is participle " Howard ", directly " Warcraft Howard " searched for to participle as one.

It should be noted that, in step S201, can carry out participle operation with reference to above-mentioned factor.

It should be noted that, the search participle that computer equipment obtains can partly overlap, and for example, search sequence " lord of the rings author Bristol gold " can have 3 search participle " lord of the rings author Bristol gold ", " lord of the rings authors " and " Bristol gold ".

It should be noted that, above-mentioned for example only for technical scheme of the present invention is described better, but not limitation of the present invention, those skilled in the art should understand that, any described search sequence is carried out to participle, obtain the implementation of described at least one search participle, all should be within the scope of the present invention.

It should be noted that, above-mentioned lifted multiple optimal way is only for illustrating better technical scheme of the present invention, but not limitation of the present invention, those skilled in the art should understand that, any implementation by obtaining at least one search participle in search sequence, as directly by search sequence is mated with a predetermined dictionary, and in search sequence can with this predetermined dictionary in the phrase of terminology match directly as search participle etc., all should be within the scope of the present invention.

In step S3, computer equipment at least one search participle based on obtaining in step S2 obtains at least one candidate personage, wherein, described at least one search participle all or part of in each search participle all can with one or more coupling the in a candidate personage's people tag.For example, the each search participle in each search participle of search sequence all can obtain coupling in any one candidate personage's people tag; Again for example, the important search participle in each search participle of search sequence all can obtain coupling etc. in any one candidate personage's people tag.Wherein, a search participle can mate with one or more people tag.

Wherein, described candidate personage can be actual nature person, as a star; Also can be virtual portrait, as a corporate body.

Particularly, computer equipment can adopt various ways, obtains at least one candidate personage.For example, computer equipment can first obtain important search participle from each search participle, then based on important search participle, determines at least one candidate personage; Wherein, each important search participle all can with one or more coupling the in any one candidate personage's people tag.Wherein, computer equipment obtains the mode of important search participle, with aforementioned to obtain the mode of important participle with reference to step S202 Computer equipment same or similar, does not repeat them here.

As a kind of optimal way of step S3, step S3 further comprises step S31 and step S32.

In step S31, computer equipment obtains personage's list of mating with the each search participle in described at least one search participle.

Particularly, for each search participle, computer equipment is by people tag coupling corresponding with each personage's list this search participle, and personage's list corresponding to the people tag that can mate, personage's list of mating as this search participle.Preferably, when a search participle can match with multiple people tag, select the highest corresponding personage's list of people tag of matching degree, personage's list of mating as this search participle.Wherein, the each personage in personage's list corresponding to people tag, all has this people tag; For example, if having 245 personages in the corresponding personage's list of people tag " doctor ", the each personage in these 245 personages, all has people tag " doctor ".

For example, computer equipment obtains 2 search participles " surgery " and " doctor " of search sequence in step S2; , for search participle " doctor ", computer equipment mates people tag corresponding with each personage's list this search participle, obtains the personage list corresponding with people tag " doctor ", personage's list of mating as search participle " doctor "; For search participle " surgery ", computer equipment is carried out above-mentioned steps equally, obtains personage's list that it mates.

It should be noted that, above-mentioned for example only for technical scheme of the present invention is described better, but not limitation of the present invention, those skilled in the art should understand that, any implementation of obtaining personage's list of mating with the each search participle in described at least one search participle, all should be within the scope of the present invention.

In step S32, computer equipment is determined the personage who all exists in the personage's list obtaining in all couplings, as described at least one candidate personage.

For example, computer equipment obtains search participle " surgery " and " doctor " in step S2, in step S31, computer equipment obtains personage's list List1 and personage's list List2 corresponding to search participle " doctor " that search participle " surgery " is corresponding, wherein, in List1, comprise personage A, B, C, D and E, in List2, comprise personage A, B, E, F and G, computer equipment is determined and in List1 and List2, is all had personage A, B and E, and the candidate personage using personage A, B and E as search sequence.

It should be noted that, above-mentioned for example only for technical scheme of the present invention is described better, but not limitation of the present invention, those skilled in the art should understand that, anyly determine the personage who all exists in the personage's list obtaining in all couplings, as described at least one candidate personage's implementation, all should be within the scope of the present invention.

In step S4, computer equipment, according to the weight information of people tag each candidate personage, that mate with described search participle, sorts to all or part of candidate personage in described at least one candidate personage.Wherein, the weight information of candidate personage's people tag can be by manually specifying, or, can be determined by computer equipment; And, by computer equipment, being determined the step of weight information, in the process that can carry out at step S4, carry out, also can before step S1, carry out.Wherein, computer equipment determines that the mode of the weight information of candidate personage's people tag will be described in detail in follow-up preferred version, at this, is not repeating.Wherein, for a candidate personage, it is higher with the weight information of the people tag that search participle mates, and its higher possibility that sorts is higher.

Particularly, computer equipment can adopt various ways according to the weight information of people tag each candidate personage, that mate with described search participle, and all or part of candidate personage in described at least one candidate personage is sorted.

As in an optimal way of step S4, step S4 further comprises step S401 and step S402.

In step S401, computer equipment is to each the candidate personage in described at least one candidate personage, according to the weight information of people tag this candidate personage, that mate with described at least one search participle, determine the weight information that this candidate personage is corresponding with described search sequence.

For example, the search participle that computer equipment obtains in step S2 is " surgery " and " doctor ", obtains candidate personage A, B and E in step S3, and wherein, candidate personage A, B and E all have people tag " surgery " and " doctor ".Candidate personage A, the B that computer equipment obtains and the weight information of E are as follows: the weighted value of the people tag " surgery " of candidate personage A is WA1, and the weighted value of people tag " doctor " is WA2; The weighted value of the people tag " surgery " of candidate personage B is WB1, and the weighted value of people tag " doctor " is WB2; The weighted value of the people tag " surgery " of candidate personage E is WE1, and the weighted value of people tag " doctor " is WE2.Computer equipment is determined according to above-mentioned weight information: weight information=WA1+WA2 that candidate personage A is corresponding with search sequence; Weight information=WB1+WB2 that candidate personage B is corresponding with search sequence; Weight information=WE1+WE2 that candidate personage E is corresponding with search sequence.

It should be noted that, above-mentioned for example only for technical scheme of the present invention is described better, but not limitation of the present invention, those skilled in the art should understand that, any to each the candidate personage in described at least one candidate personage, according to the weight information of people tag this candidate personage, that mate with described at least one search participle, determine the implementation of this candidate personage weight information corresponding with described search sequence, all should be within the scope of the present invention.

In step S402, computer equipment, according to each candidate personage's weight information, sorts to all or part of candidate personage in described at least one candidate personage.

For example, computer equipment obtains candidate personage A, B and E in step S3.Computer equipment determines that in step S401 the weight of candidate personage A is 95, and the weight of candidate personage B is 85, and the weight of candidate personage E is 92.In step S402, computer equipment according to above-mentioned weight information to candidate personage according to weight from height to low sequence, obtain ranking results and be: candidate personage A, candidate personage E, candidate personage B.

It should be noted that, computer equipment can, according to each candidate personage's weight information, sort to all or part of candidate personage in described at least one candidate personage according to pre-defined rule.Wherein, described pre-defined rule is for determining the number that can participate in the candidate personage of sequence or can participate in the candidate personage of sequence.

For example, pre-defined rule comprises that the candidate personage that weight in candidate personage is exceeded to predefined weight threshold value sorts, and predefined weight threshold value is 90, computer equipment obtains candidate personage A, B and E in step S3, and the weight of determining candidate personage A in step S401 is 95, the weight of candidate personage B is 85, and the weight of candidate personage E is 92.In step S402, computer equipment sorts to the candidate personage A and the E that meet this pre-defined rule, obtains ranking results to be: candidate personage A, candidate personage E.

Again for example, pre-defined rule comprises that the quantity of the thing that stands for election exceedes at 100 o'clock, only 100 candidate personages is wherein sorted.Computer equipment can first be determined the candidate personage of first 100 of weight, then these 100 candidate personages are sorted etc.

It should be noted that, above-mentioned for example only for technical scheme of the present invention is described better, but not limitation of the present invention, those skilled in the art should understand that, any according to each candidate personage's weight information, the implementation that all or part of candidate personage in described at least one candidate personage is sorted, all should be within the scope of the present invention.

It should be noted that, step S401 and step S402 there is no strict sequencing, computer equipment can perform step S402 in execution step S401, for example, computer equipment can obtain after a candidate personage's weight information, according to itself and other candidate personage's weight information, be inserted into suitable sorting position, meanwhile, computer equipment also performs step S401, obtains another candidate personage's weight information.

And for example, in another optimal way of step S4, step S4 further comprises step S4 '.

In step S4 ', computer equipment is according to the weight information of people tag each candidate personage, that mate with described search participle, and in conjunction with authentication information corresponding to each candidate personage, all or part of candidate personage in described at least one candidate personage is sorted.

Wherein, for a candidate personage, it is higher with the weight information of the people tag that search participle mates, and its higher possibility that sorts is higher; It has authentication information, and its higher possibility that sorts is higher.

Particularly, computer equipment can adopt various ways, come according to the weight information of people tag each candidate personage, that mate with described search participle, and in conjunction with authentication information corresponding to each candidate personage, all or part of candidate personage in described at least one candidate personage is sorted.

For example, for each candidate personage, computer equipment can be directly based on the weight information of people tag and this candidate personage's authentication information this candidate personage, that mate with search participle, calculate this candidate personage weight corresponding with search sequence, and, the weight that computer equipment is corresponding with this search sequence according to each candidate personage, to all or part of sequence the in each candidate personage.

And step S4 ' can combine with step S401 and S402.For example: to each the candidate personage in described at least one candidate personage, computer equipment is according to the weight information of people tag this candidate personage, that mate with described at least one search participle, determine the weight information that this candidate personage is corresponding with described search sequence, and according to each candidate personage's weight information, all or part of candidate personage in described at least one candidate personage is sorted; Then, whether computer equipment has authentication information according to each candidate personage, improves the sequence of the candidate personage with authentication information, and reduces the sequence of the candidate personage without authentication information.Again for example, to each the candidate personage in described at least one candidate personage, computer equipment is according to the weight information of people tag this candidate personage, that mate with described at least one search participle, determine the weight information that this candidate personage is corresponding with described search sequence, and, whether computer equipment has authentication information according to each candidate personage, and each candidate personage's weight information is weighted; And computer equipment, according to the weight information after weighting, sorts to all or part of candidate personage at least one candidate personage etc.

It should be noted that, above-mentioned for example only for technical scheme of the present invention is described better, but not limitation of the present invention, those skilled in the art should understand that, any according to the weight information of people tag each candidate personage, that mate with described search participle, and in conjunction with authentication information corresponding to each candidate personage, the implementation that all or part of candidate personage in described at least one candidate personage is sorted, all should be within the scope of the present invention.

It should be noted that, step S3 and step S4 there is no strict sequencing, and when computer equipment can and obtain candidate personage at execution step S3, execution step S4 sorts to acquired candidate personage.

In step S5, computer equipment offers described user by the candidate personage after sequence.

Preferably, when computer equipment offers described user by the candidate personage after sequence, the additional information in candidate personage can be offered to user in the lump, wherein, described additional information can comprise this candidate personage's descriptor.

It should be noted that, above-mentioned for example only for technical scheme of the present invention is described better, but not limitation of the present invention those skilled in the art should understand that, any implementation that candidate personage after sequence is offered to described user, all should be within the scope of the present invention.

As a kind of preferred version of the present embodiment, computer equipment, according to following at least one factor, is determined the weight information of personage's people tag:

1) degree of concern of other personages that there is this people tag to this personage.

Other personages wherein, with this people tag can determine based on following factor this personage's degree of concern: other personages with this people tag pay close attention to or forward the quantity of the content of this personage's issue, pay close attention to this personage and have other personages' of this people tag quantity etc.

For example, personage A has " lawyer " label, can be according to paying close attention to the quantity in other personages of personage A with the personage of " lawyer " label, and/or, the personage with " lawyer " label forwards the quantity of the notice of this personage A issue etc., determines the weight information of " lawyer " label of personage A.

2) this personage is to having other personages' the degree of concern of this people tag.

Wherein, this personage can determine based on following factor the degree of concern of other personages with this people tag: the quantity of the content of other personages' with this people tag that this personage pays close attention to quantity, this personage other personage's issues that forward, that have this people tag.

For example, personage A has " lawyer " label, in other personages that pay close attention to according to personage A, there is the personage's of " lawyer " label quantity, and/or, the quantity of the content of personage A other personages issue that forward, that have " lawyer " label, determines the weight information of " lawyer " label of personage A.

3) quantity of the content of this personage feature issue, that there is this people tag.

Wherein, computer equipment can determine whether this content has the feature of this people tag by the content of resolving this personage's issue.

For example, can, by resolving the phrase etc. that whether comprises the feature with this people tag in the content of this personage issue, determine whether this content has the feature of this people tag.

4) degree that the content of this personage feature issue, that have this people tag is paid close attention to by other personages.Wherein, can determine whether this content has the feature of this people tag by the content of analyzing personage's issue.

For example, according to the content of feature personage A issue, that there is " lawyer " label, by other personages, forwarded or the quantity of mark, determine the weight information of " lawyer " label of personage A.

5) degree of concern of the content of this personage to other personages feature issue, that there is this people tag.

For example, according to personage A, other personages content issue, that have the feature of " lawyer " label is forwarded or the quantity of mark, determine the weight information of " lawyer " label of personage A.

6) whether this personage has the authentication information corresponding with this people tag.

Wherein, this authentication information comprises the information of any authenticity that can be used in this personage of authentication.As authentication information can include but not limited to: professionally certified information, as lawyer demonstrate,proves information, doctor's operation card information, teacher's certificate information etc.; Work unit's information under this personage, as lawyer's office's information at the information for hospital at doctor place, lawyer place etc.; Authentication information, as real name authentication information etc.Preferably, a personage can have multiple authentication informations, and the people tag that each authentication information can be corresponding different,, when this personage is sorted as candidate personage, can only consider the authentication information corresponding with the people tag of search participle coupling.

It should be noted that, above-mentioned for example only for technical scheme of the present invention is described better, but not limitation of the present invention those skilled in the art should understand that, the implementation of the weight information of any definite personage's people tag, all should be within the scope of the present invention.

According to the scheme of the present embodiment, can retrieve for personage, and the weight of people tag that can be based on mating with search sequence, candidate personage is sorted, thereby make the fiduciary level of personage's result for retrieval higher; Preferably, the present embodiment can be determined the weight information of its people tag in conjunction with personage's community network activity or personal information, thereby has further guaranteed search based on people tag and the accuracy of sequence; Preferably, the scheme of the present embodiment can realize personage's search in universal search engine, thereby can realize personage's function of search of across a network community.

Fig. 2 be another embodiment of the present invention for searching for personage's method flow schematic diagram.

The method of the present embodiment is mainly realized by computer equipment; Wherein, to any explanation with reference to the computer equipment described in embodiment illustrated in fig. 1, be all contained in by reference in the present embodiment.Wherein, the method for the present embodiment comprises step S1, step S2, step S3, step S4, step S5 and step S6.Wherein, step S1, step S2, step S3, step S4 and step S5 are described in detail in reference to Fig. 1, do not repeat them here.

In step S6, computer equipment, according to the content of other personages' that are associated with a personage people tag and/or described other personage's issues, upgrades described personage's people tag.

Wherein, described other personages that are associated with a personage comprise other personages that have cyberrelationship with this personage; For example, this personage pays close attention to other personages, pay close attention to other personages of this personage, there are other personages etc. of good friend's relation with this personage.

Wherein, the renewal of described people tag includes but not limited to increase, modification and the deletion etc. of people tag.

Particularly, computer equipment is according to the content of other personages' that are associated with a personage people tag and/or described other personage's issues, and the mode of upgrading described personage's people tag includes but not limited to:

1) computer equipment, directly according to the content of other personages' that are associated with a personage people tag and/or described other personage's issues, obtains one or more people tag, and according to obtained people tag, upgrades this personage's people tag.

Wherein, computer equipment can adopt various ways, comes, according to the content of other personages' that are associated with a personage people tag and/or described other personage's issues, to obtain one or more people tag.

For example, computer equipment is added up other personages' that are associated with a personage people tag, and by the quantity higher sequence that occurs quantity or that occur higher than the people tag of a predetermined quantity, as the people tag that is applicable to this personage.

Again for example, for other personages' people tag Tag1, computer equipment is added up the number of times that this people tag Tag1 occurs in all other personages' associated with this figure picture people tag, and the number of times occurring in the people tag in a large amount of personages (as the personage storehouse of computer equipment) in conjunction with this people tag Tag1, determine whether this people tag Tag1 can characterize other personages' associated with this figure picture feature, and people tag that can characteristic feature is as the people tag that is applicable to described renewal.Wherein, the number of times that people tag Tag1 occurs in other personages' associated with this figure picture people tag is higher, and the number of times occurring in the people tag in a large amount of personages (as the personage storehouse of computer equipment) is lower, the characteristic present ability of people tag Tag1 is stronger, is more applicable to upgrade operation; The number of times that people tag Tag1 occurs in other personages' associated with this figure picture people tag is lower, and the number of times occurring in the people tag in a large amount of personages (as the personage storehouse of computer equipment) is higher,, a little less than the characteristic present sexuality of people tag Tag1, be not suitable for and upgrade operation.

Again for example, computer equipment is according to the content of other personage's issues that are associated with a personage, and the feature of the people tag having according to these contents, determines one or more people tag.In the content that other personages that are associated according to a personage as computer equipment issue, there is a large amount of medical speciality vocabulary, determine people tag " doctor " etc.

2) step S6 further comprises step S61, step S62 and step S63.

In step S61, computer equipment, according to other personages' associated with described figure picture address feature and network social intercourse relationship characteristic, is divided into multiple groupings by other personages associated with described figure picture.

Preferably, described address feature comprises network address feature.For example, comprise the IP address that other personages use.

Preferably, described network social intercourse relation can comprise any incidence relation in Web Community.For example, other personages exist concern relation, other personages and this personage to exist good friend's relation, other personages and this personage to pay close attention to identical personage etc. with this personage.

Particularly, there are other personages of same or analogous address feature or network social intercourse relationship characteristic, be more easily divided into one group.For example, other personages of identical IP address or close (as front in IP address 3 sections identical) are divided into one group; Again for example, will have concern relation or good friend's relation with this personage, and other personages that all pay close attention to same star personality are divided into one group etc., wherein, star personality can determine factors such as bean vermicelli number by it.

It should be noted that, above-mentioned for example only for technical scheme of the present invention is described better, but not limitation of the present invention, those skilled in the art should understand that, any basis other personages' associated with described figure picture address feature and network social intercourse relationship characteristic, other personages associated with described figure picture are divided into the implementation of multiple groupings, all should be within the scope of the present invention.

In step S62, computer equipment, according to the content of described other personages' that in described multiple groupings, all or part of grouping comprises people tag and/or described other personage's issues, obtains one or more people tag.Wherein, the content of described other personages' that computer equipment comprises according to each grouping people tag and/or described other personage's issues, obtain the mode of one or more people tag, aforementioned implementation 1 with step S6) in, computer equipment is directly according to the content of other personages' that are associated with a personage people tag and/or described other personage's issues, the mode that obtains one or more people tag is same or similar, does not repeat them here.

For example, in step S61, computer equipment is divided into group G1 and group G2 by other personages that are associated with personage A, in step S62, computer equipment is resolved and is obtained organizing in G1 90% personage and all have " doctor " label, and the content 80% of the personage's issue in group G1 is relevant with " surgery ", and computer equipment obtains " doctor " label and " surgery " label.

It should be noted that, above-mentioned for example only for technical scheme of the present invention is described better, but not limitation of the present invention, those skilled in the art should understand that, any according to the content of described other personages' that in described multiple groupings, all or part of grouping comprises people tag and/or described other personage's issues, obtain the implementation of one or more people tag, all should be within the scope of the present invention.

In step S63, computer equipment, according to described one or more people tag, upgrades described personage's people tag.

For example, computer equipment obtains " doctor " label and " surgery " label according to other personages that are associated with personage A in step S62, and personage A does not have this two labels, for personage A adds this two labels.

It should be noted that, above-mentioned for example only for technical scheme of the present invention is described better, but not limitation of the present invention, those skilled in the art should understand that, any according to described one or more people tag, upgrade the implementation of described personage's people tag, all should be within the scope of the present invention.

It should be noted that, above-mentioned for example only for technical scheme of the present invention is described better, but not limitation of the present invention, those skilled in the art should understand that, the content of other personages' that any basis is associated with a personage people tag and/or described other personage's issues, upgrade the implementation of described personage's people tag, all should be within the scope of the present invention.

Follow the scheme according to the present embodiment, by other personages associated with figure picture, upgrade label for this personage, make the reaction personage's that personage's label can be more effective, more complete feature, and the operations such as interpolation, deletion and the change by people tag make people tag can react more accurately the feature that personage is current had, reduced in personage's search and occurred that personage omits or the inaccurate situation of people tag, thereby the Search Results that personage is searched for is more comprehensive, also more meet user's search condition.

It should be noted that, for simplicity, before step S6 being placed in to step S1 in Fig. 2, but those skilled in the art will be understood that step S6 and step S1, there is no sequencing to step S5.

Fig. 3 be another embodiment of the present invention for searching for personage's method flow schematic diagram.

The method of the present embodiment is mainly realized by computer equipment; Wherein, to any explanation with reference to the computer equipment described in embodiment illustrated in fig. 1, be all contained in by reference in the present embodiment.Wherein, the method for the present embodiment comprises step S1, step S2, step S3, step S4, step S5, step S7 and step S8.Wherein, step S1, step S2, step S3, step S4 and step S5 are described in detail in reference to Fig. 1, do not repeat them here.

In step S7, computer equipment, according to the content of described personage's issue, is determined at least one people tag.

Particularly, computer equipment, according to the content of personage's issue, is determined the mode of at least one people tag, according to the content of other personage's issues, determines that the mode of at least one people tag is same or similar with abovementioned steps S6 Computer equipment.

For example, each in the content that computer equipment is issued personage A is resolved, extract the frequency that each word in this content occurs, get rid of everyday words, modal particle and conjunction etc., obtain the keyword of the most much higher word of the frequency of occurrences as this content, and judge that by semantic analysis the people tag that these keywords are corresponding is " doctor " label.

It should be noted that, above-mentioned for example only for technical scheme of the present invention is described better, but not limitation of the present invention, those skilled in the art should understand that, according to the content of described personage's issue, determine the implementation of at least one people tag, all should be within the scope of the present invention.

In step S8, computer equipment, according to described at least one people tag, upgrades described personage's people tag.

Wherein, computer equipment, according to described at least one people tag, upgrades the mode of described personage's people tag, with abovementioned steps S63 Computer equipment according to one or more people tag, the mode of people tag of upgrading described personage is same or similar, does not repeat them here.

It should be noted that, for simplicity, before step S7 and S8 being placed in to step S1 in Fig. 3, but those skilled in the art will be understood that step S7 and S8 and step S1, there is no sequencing to step S5.

Follow the scheme according to the present embodiment, personage's people tag can be upgraded according to the content distributed of personage, make the reaction personage's that personage's label can be more effective, more complete feature, and can be by the content of personage self issue, react more accurately the feature that personage is current had, reduced in personage's search and occurred that personage omits or the inaccurate situation of people tag, thereby the Search Results that personage is searched for is more comprehensive, also more meets user's search condition.

Fig. 4 be one embodiment of the invention for searching for personage's the structural representation of searcher.

The searcher of the present embodiment comprises the first acquisition device 1, the second acquisition device 2, the first determining device 3, collator 4 and generator 5.Preferably, the searcher of the present embodiment can be included in computer equipment.This computer equipment can be realized the function of universal search engine, and this universal search engine can obtain the people information of across a network community.

The first acquisition device 1 obtains the search sequence from user.

Wherein, the first acquisition device 1 obtains from the mode of user's search sequence and includes but not limited to:

1) computer equipment at searcher place is the subscriber equipment of direct and user interactions, and the first acquisition device 1 directly obtains the search sequence of user's input.

For example, the computer equipment at searcher place is PC, and the first acquisition device 1 directly obtains the search sequence that user directly inputs in the browser searches input frame of PC.

2) the first acquisition device 1 obtains the search sequence of user's input by other computer equipments.

For example, the computer equipment at searcher place is the network equipment, user is inputted search sequence in the browser searches input frame of subscriber equipment, and subscriber equipment offers the network equipment by this search sequence by network, and the first acquisition device 1 is by this search sequence of Network Capture.

The second acquisition device 2 is by obtaining at least one search participle in described search sequence.

Particularly, the second acquisition device 2 can adopt various ways, obtains at least one search participle from search sequence.

As a kind of optimal way as the second acquisition device 2, the second acquisition device 2 further comprises participle device (scheming not shown) and selecting arrangement (scheming not shown).

Participle device carries out participle to described search sequence, obtains participle list.

For example, the first acquisition device 1 obtains the search sequence " lawyer of economic class " from user, participle device, according to semantic analysis and predetermined participle dictionary, carries out participle to this search sequence, obtain participle list comprise " economic class ", " " and the participle list of " lawyer ".

Selecting arrangement is selected important participle from described participle list, as described at least one search participle.Wherein, selecting arrangement can determine whether the participle in participle list is important participle in several ways; For example selecting arrangement can be by mating the participle in participle list with the participle in predetermined participle dictionary, and/or, semantic analysis, the semantic analysis to the participle in participle list itself and the syntactic position analysis of the participle based in participle list in search sequence etc. based on to search sequence, determine whether the participle in participle list is important participle.

For example, participle device is by search sequence " lawyer of economic class " is carried out to participle, obtains participle list, the participle wherein comprising be " economic class ", " " and " lawyer "; Selecting arrangement mates the participle in participle list with the participle in predetermined participle dictionary, and selects important participle " economic class " and " lawyer " according to matching result, as corresponding two the search participles of search sequence " lawyer of economic class ".

And for example, as the another kind of optimal way of the second acquisition device 2, the second acquisition device 2 comprises sub-acquisition device (scheming not shown).

Sub-acquisition device carries out participle to described search sequence, obtains described at least one search participle.

Wherein, the operation of the participle that sub-acquisition device carries out search sequence, consider following at least one factor:

Preferably, sub-acquisition device is selected the less participle mode of semantic ambiguity.If there is the subphrase larger with the semantic differential of this phrase in multiple subphrases that a phrase in search sequence obtains after being further divided, select this complete phrase as a participle, and do not select the subphrase of this phrase as participle.When and for example in search sequence, multiple phrases that do not comprise mutually can produce multiple combination, after Selection and Constitute with the less array mode of search sequence semantic difference etc.It should be noted that, search sequence can be directly a phrase.

For example, search sequence comprises phrase " i Phone ", multiple subphrases that this phrase obtains after being further divided are " apple " and " mobile phone ", sub-acquisition device is by the pre-definite phrase classification of inquiry, the primary categories of determining phrase " i Phone " is " mobile phone ", and the primary categories of determining the subphrase " apple " of this phrase comprises " fruit " and " mobile phone ", the primary categories of subphrase " mobile phone " is " mobile phone ", in multiple subphrases that sub-acquisition device judgement " i Phone " obtains after being further divided, there is the subphrase " apple " larger with the semantic differential of this phrase, sub-acquisition device selects phrase " i Phone " as a participle, and do not select the subphrase " apple " of this phrase and " mobile phone " as participle.

It should be noted that, participle device can carry out participle operation with reference to above-mentioned factor.

It should be noted that, the search participle that searcher obtains can partly overlap, and for example, search sequence " lord of the rings author Bristol gold " can have 3 search participle " lord of the rings author Bristol gold ", " lord of the rings authors " and " Bristol gold ".

At least one search participle that the first determining device 3 obtains based on the second acquisition device 2 obtains at least one candidate personage, wherein, described at least one search participle all or part of in each search participle all can with one or more coupling the in a candidate personage's people tag.For example, the each search participle in each search participle of search sequence all can obtain coupling in any one candidate personage's people tag; Again for example, the important search participle in each search participle of search sequence all can obtain coupling etc. in any one candidate personage's people tag.Wherein, a search participle can mate with one or more people tag.

Particularly, the first determining device 3 can adopt various ways, obtains at least one candidate personage.For example, the first determining device 3 can first obtain important search participle from each search participle, then based on important search participle, determines at least one candidate personage; Wherein, each important search participle all can with one or more coupling the in any one candidate personage's people tag.Wherein, the first determining device 3 obtains the mode of important search participle, and the mode that obtains important participle with aforementioned selecting arrangement is same or similar, does not repeat them here.

As a kind of optimal way of the first determining device 3, the first determining device 3 further comprises list of steps acquisition device (scheming not shown) and personage's determining device (scheming not shown).

List acquisition device obtains personage's list of mating with the each search participle in described at least one search participle.

Particularly, for each search participle, list acquisition device is by people tag coupling corresponding with each personage's list this search participle, and personage's list corresponding to the people tag that can mate, personage's list of mating as this search participle.Preferably, when a search participle can match with multiple people tag, select the highest corresponding personage's list of people tag of matching degree, personage's list of mating as this search participle.Wherein, the each personage in personage's list corresponding to people tag, all has this people tag; For example, if having 245 personages in the corresponding personage's list of people tag " doctor ", the each personage in these 245 personages, all has people tag " doctor ".

For example, the second acquisition device 2 obtains 2 search participles " surgery " and " doctor " of search sequence; , for search participle " doctor ", list acquisition device mates people tag corresponding with each personage's list this search participle, obtains the personage list corresponding with people tag " doctor ", personage's list of mating as search participle " doctor "; For search participle " surgery ", list acquisition device is carried out aforesaid operations equally, obtains personage's list that it mates.

Personage's determining device is determined the personage who all exists in the personage's list obtaining in all couplings, as described at least one candidate personage.

For example, the second acquisition device 2 obtains search participle " surgery " and " doctor ", list acquisition device obtains personage's list List1 and personage's list List2 corresponding to search participle " doctor " that search participle " surgery " is corresponding, wherein, in List1, comprise personage A, B, C, D and E, in List2, comprise personage A, B, E, F and G, personage's determining device is determined and in List1 and List2, is all had personage A, B and E, and the candidate personage using personage A, B and E as search sequence.

Collator 4, according to the weight information of people tag each candidate personage, that mate with described search participle, sorts to all or part of candidate personage in described at least one candidate personage.Wherein, the weight information of candidate personage's people tag can be by manually specifying, or, can be determined by computer equipment; And, by computer equipment, being determined the step of weight information, can in the process of collator 4 executable operations, carry out, also can before the first acquisition device 1, carry out.Wherein, computer equipment determines that the mode of the weight information of candidate personage's people tag will be described in detail in follow-up preferred version, at this, is not repeating.Wherein, for a candidate personage, it is higher with the weight information of the people tag that search participle mates, and its higher possibility that sorts is higher.

Particularly, collator 4 can adopt various ways according to the weight information of people tag each candidate personage, that mate with described search participle, and all or part of candidate personage in described at least one candidate personage is sorted.

As in an optimal way of collator 4, collator 4 further comprises the first weight determining device (scheming not shown) and the first sub-collator (scheming not shown).

The first weight determining device is to each the candidate personage in described at least one candidate personage, according to the weight information of people tag this candidate personage, that mate with described at least one search participle, determine the weight information that this candidate personage is corresponding with described search sequence.

For example, the search participle that the second acquisition device 2 obtains is " surgery " and " doctor ", and the first determining device 3 obtains candidate personage A, B and E, and wherein, candidate personage A, B and E all have people tag " surgery " and " doctor ".Candidate personage A, the B that the first weight determining device obtains and the weight information of E are as follows: the weighted value of the people tag " surgery " of candidate personage A is WA1, and the weighted value of people tag " doctor " is WA2; The weighted value of the people tag " surgery " of candidate personage B is WB1, and the weighted value of people tag " doctor " is WB2; The weighted value of the people tag " surgery " of candidate personage E is WE1, and the weighted value of people tag " doctor " is WE2.The first weight determining device is determined according to above-mentioned weight information: weight information=WA1+WA2 that candidate personage A is corresponding with search sequence; Weight information=WB1+WB2 that candidate personage B is corresponding with search sequence; Weight information=WE1+WE2 that candidate personage E is corresponding with search sequence.

The first sub-collator, according to each candidate personage's weight information, sorts to all or part of candidate personage in described at least one candidate personage.

For example, the first determining device 3 obtains candidate personage A, B and E.The first weight determining device determines that the weight of candidate personage A is 95, and the weight of candidate personage B is 85, and the weight of candidate personage E is 92.The first sub-collator according to above-mentioned weight information to candidate personage according to weight from height to low sequence, obtain ranking results and be: candidate personage A, candidate personage E, candidate personage B.

It should be noted that, the first sub-collator can, according to each candidate personage's weight information, sort to all or part of candidate personage in described at least one candidate personage according to pre-defined rule.Wherein, described pre-defined rule is for determining the number that can participate in the candidate personage of sequence or can participate in the candidate personage of sequence.

For example, pre-defined rule comprises that the candidate personage that weight in candidate personage is exceeded to predefined weight threshold value sorts, and predefined weight threshold value is 90, the first determining device 3 obtains candidate personage A, B and E, the first weight determining device determines that the weight of candidate personage A is 95, the weight of candidate personage B is 85, and the weight of candidate personage E is 92.The first sub-collator sorts to the candidate personage A and the E that meet this pre-defined rule, obtains ranking results to be: candidate personage A, candidate personage E.

Again for example, pre-defined rule comprises that the quantity of the thing that stands for election exceedes at 100 o'clock, only 100 candidate personages is wherein sorted.The first sub-collator can first be determined the candidate personage of first 100 of weight, then these 100 candidate personages are sorted etc.

It should be noted that, the operation that the triggering of the first weight determining device and the first sub-collator is carried out there is no strict sequencing, the first weight determining device can with the first sub-collator executable operations simultaneously, for example, searcher is obtaining after a candidate personage's weight information, and the first sub-collator according to itself and other candidate personage's weight information, is inserted into suitable sorting position, meanwhile, the first weight determining device is obtained another candidate personage's weight information.

And for example, in another optimal way of collator 4, collator 4 further comprises the second sub-collator (scheming not shown).

The second sub-collator is according to the weight information of people tag each candidate personage, that mate with described search participle, and in conjunction with authentication information corresponding to each candidate personage, all or part of candidate personage in described at least one candidate personage is sorted.

Particularly, the second sub-collator can adopt various ways, come according to the weight information of people tag each candidate personage, that mate with described search participle, and in conjunction with authentication information corresponding to each candidate personage, all or part of candidate personage in described at least one candidate personage is sorted.

For example, for each candidate personage, the second sub-collator can be directly based on the weight information of people tag and this candidate personage's authentication information this candidate personage, that mate with search participle, calculate this candidate personage weight corresponding with search sequence, and, the weight that the second sub-collator is corresponding with this search sequence according to each candidate personage, to all or part of sequence the in each candidate personage.

And the second sub-collator can combine with the function that the first weight determining device and the first sub-collator are realized.For example: to each the candidate personage in described at least one candidate personage, and according to the weight information of people tag this candidate personage, that mate with described at least one search participle, determine the weight information that this candidate personage is corresponding with described search sequence, and according to each candidate personage's weight information, all or part of candidate personage in described at least one candidate personage is sorted; Then, according to each candidate personage, whether there is authentication information, improve the sequence of the candidate personage with authentication information, and reduce the sequence of the candidate personage without authentication information.Again for example, to each the candidate personage in described at least one candidate personage, can be according to the weight information of people tag this candidate personage, that mate with described at least one search participle, determine the weight information that this candidate personage is corresponding with described search sequence, and, according to each candidate personage, whether there is authentication information, each candidate personage's weight information is weighted; And, according to the weight information after weighting, all or part of candidate personage at least one candidate personage is sorted etc.

It should be noted that, the operation that the first determining device 3 and collator 4 are carried out there is no strict sequencing, and in the first determining device 3 executable operations and when obtaining candidate personage, collator 4 can sort to acquired candidate personage.

Generator 5 offers described user by the candidate personage after sequence.

Preferably, when generator 5 offers described user by the candidate personage after sequence, the additional information in candidate personage can be offered to user in the lump, wherein, described additional information can comprise this candidate personage's descriptor.

As a kind of preferred version of the present embodiment, searcher comprises the second weight determining device (scheming not shown), and the second weight determining device, for following at least one the factor of basis, is determined the weight information of personage's people tag:

Wherein, the second weight determining device can determine whether this content has the feature of this people tag by the content of resolving this personage's issue.

Fig. 5 be another embodiment of the present invention for searching for personage's the result schematic diagram of searcher.

Wherein, the searcher of the present embodiment comprises the first acquisition device 1, the second acquisition device 2, the first determining device 3, collator 4, generator 5 and the first updating device 6.Wherein, the first acquisition device 1, the second acquisition device 2, the first determining device 3, collator 4 and generator 5 have been described in detail in reference to Fig. 4, do not repeat them here.

The first updating device 6, according to the content of other personages' that are associated with a personage people tag and/or described other personage's issues, upgrades described personage's people tag.

Particularly, the first updating device 6 is according to the content of other personages' that are associated with a personage people tag and/or described other personage's issues, and the mode of upgrading described personage's people tag includes but not limited to:

1) the first updating device 6, directly according to the content of other personages' that are associated with a personage people tag and/or described other personage's issues, obtains one or more people tag, and according to obtained people tag, upgrades this personage's people tag.

Wherein, the first updating device 6 can adopt various ways, comes, according to the content of other personages' that are associated with a personage people tag and/or described other personage's issues, to obtain one or more people tag.

For example, the people tag that the first updating device 6 is added up other personages that are associated with a personage, and by the quantity higher sequence that occurs quantity or that occur higher than the people tag of a predetermined quantity, as the people tag that is applicable to this personage.

Again for example, for other personages' people tag Tag1, the number of times that first updating device 6 this people tag of statistics Tag1 occur in all other personages' associated with this figure picture people tag, and the number of times occurring in the people tag in a large amount of personages (as the personage storehouse of computer equipment) in conjunction with this people tag Tag1, determine whether this people tag Tag1 can characterize other personages' associated with this figure picture feature, and people tag that can characteristic feature is as the people tag that is applicable to described renewal.Wherein, the number of times that people tag Tag1 occurs in other personages' associated with this figure picture people tag is higher, and the number of times occurring in the people tag in a large amount of personages (as the personage storehouse of computer equipment) is lower, the characteristic present ability of people tag Tag1 is stronger, is more applicable to upgrade operation; The number of times that people tag Tag1 occurs in other personages' associated with this figure picture people tag is lower, and the number of times occurring in the people tag in a large amount of personages (as the personage storehouse of computer equipment) is higher,, a little less than the characteristic present sexuality of people tag Tag1, be not suitable for and upgrade operation.

Again for example, the first updating device 6 is according to the content of other personage's issues that are associated with a personage, and the feature of the people tag having according to these contents, determines one or more people tag.In the content that other personages that are associated according to a personage as the first updating device 6 issue, there is a large amount of medical speciality vocabulary, determine people tag " doctor " etc.

2) the first updating device 6 further comprises apparatus for grouping (scheming not shown), the first label determining device (scheming not shown) and sub-updating device (scheming not shown).

Apparatus for grouping, according to other personages' associated with described figure picture address feature and network social intercourse relationship characteristic, is divided into multiple groupings by other personages associated with described figure picture.

The first label determining device, according to the content of described other personages' that in described multiple groupings, all or part of grouping comprises people tag and/or described other personage's issues, obtains one or more people tag.Wherein, the content of described other personages' that the first label determining device comprises according to each grouping people tag and/or described other personage's issues, obtain the mode of one or more people tag, aforementioned implementation 1 with the first updating device 6) in, the first updating device 6 is directly according to the content of other personages' that are associated with a personage people tag and/or described other personage's issues, the mode that obtains one or more people tag is same or similar, does not repeat them here.

For example, apparatus for grouping is divided into group G1 and group G2 by other personages that are associated with personage A, the first label determining device is resolved and is obtained organizing in G1 90% personage and all have " doctor " label, and the content 80% of the personage's issue in group G1 is relevant with " surgery ", and the first label determining device obtains " doctor " label and " surgery " label.

Sub-updating device, according to described one or more people tag, upgrades described personage's people tag.

For example, the first label determining device obtains " doctor " labels and " surgery " labels according to other personages that are associated with personage A, and personage A does not have this two labels, and sub-updating device adds this two labels for personage A.

It should be noted that, for simplicity, the logical order that shows the first updating device 6 in Fig. 5 there is no sequencing but those skilled in the art will be understood that the operation that the first updating device 6 and the first acquisition device 1 are carried out to generator 5 before being positioned at the first acquisition device 1.

The searcher of the present embodiment comprises the first acquisition device 1, the second acquisition device 2, the first determining device 3, collator 4, generator 5, the second label determining device 7 and the second updating device 8.Wherein, the first acquisition device 1, the second acquisition device 2, the first determining device 3, collator 4 and generator 5 have been described in detail in reference to Fig. 4, do not repeat them here.

The second label determining device 7, according to the content of described personage's issue, is determined at least one people tag.

Particularly, the second label determining device 7, according to the content of personage's issue, is determined the mode of at least one people tag, according to the content of other personage's issues, determines that the mode of at least one people tag is same or similar with aforementioned the first updating device 6.

For example, each in the content that the second label determining device 7 is issued personage A is resolved, extract the frequency that each word in this content occurs, get rid of everyday words, modal particle and conjunction etc., obtain the keyword of the most much higher word of the frequency of occurrences as this content, and judge that by semantic analysis the people tag that these keywords are corresponding is " doctor " label.

The second updating device 8, according to described at least one people tag, upgrades described personage's people tag.

Wherein, the second updating device 8, according to described at least one people tag, upgrades the mode of described personage's people tag, with aforementioned sub-updating device according to one or more people tag, the mode of people tag of upgrading described personage is same or similar, does not repeat them here.

It should be noted that, for simplicity, in Fig. 6, show that the logical order of the first acquisition device 1 is before the second label determining device 7 and the second updating device 8, but those skilled in the art will be understood that between the operation that the second label determining device 7 and the second updating device 8 and the first acquisition device 1 to generator 5 carry out and there is no sequencing.

It should be noted that the present invention can be implemented in the assembly of software and/or software and hardware, for example, each device of the present invention can adopt special IC (ASIC) or any other similar hardware device to realize.In one embodiment, software program of the present invention can carry out to realize step mentioned above or function by processor.Similarly, software program of the present invention (comprising relevant data structure) can be stored in computer readable recording medium storing program for performing, for example, and RAM storer, magnetic or CD-ROM driver or flexible plastic disc and similar devices.In addition, steps more of the present invention or function can adopt hardware to realize, for example, thereby as coordinate the circuit of carrying out each step or function with processor.

To those skilled in the art, obviously the invention is not restricted to the details of above-mentioned example embodiment, and in the situation that not deviating from spirit of the present invention or essential characteristic, can realize the present invention with other concrete form.Therefore, no matter from which point, all should regard embodiment as exemplary, and be nonrestrictive, scope of the present invention is limited by claims rather than above-mentioned explanation, is therefore intended to all changes that drop in the implication and the scope that are equal to important document of claim to be included in the present invention.Any Reference numeral in claim should be considered as limiting related claim.In addition, obviously other unit or step do not got rid of in " comprising " word, and odd number is not got rid of plural number.Multiple unit of stating in system claim or device also can be realized by software or hardware by a unit or device.The first, the second word such as grade is used for representing title, and does not represent any specific order.

Claims

1. for searching for personage's a method, wherein, the method comprises the following steps:

A obtains the search sequence from user;

B is by obtaining at least one search participle in described search sequence;

E offers described user by the candidate personage after sequence.

2. method according to claim 1, wherein, described step c comprises the following steps:

-described in obtaining, at least one searches for personage's list that the each search participle in participle mates;

-determine the personage who all exists in the personage's list obtaining in all couplings, as described at least one candidate personage.

3. method according to claim 1 and 2, wherein, described steps d comprises the following steps:

-to each the candidate personage in described at least one candidate personage, according to the weight information of people tag this candidate personage, that mate with described at least one search participle, determine the weight information that this candidate personage is corresponding with described search sequence;

-according to each candidate personage's weight information, all or part of candidate personage in described at least one candidate personage is sorted.

4. method according to claim 1 and 2, wherein, described steps d comprises the following steps:

-according to the weight information of people tag each candidate personage, that mate with described search participle, and in conjunction with authentication information corresponding to each candidate personage, all or part of candidate personage in described at least one candidate personage is sorted.

5. according to the method described in any one in claim 1 to 4, wherein, described step b comprises the following steps:

-described search sequence is carried out to participle, obtain participle list;

-from described participle list, select important participle, as described at least one search participle.

6. according to the method described in any one in claim 1 to 4, wherein, described step b comprises the following steps:

-described search sequence is carried out to participle, obtain described at least one search participle;

Wherein, following at least one factor considered in the operation of described participle:

The semantic difference of the different word segmentation result of-described search sequence;

The matching result of-described search sequence and predetermined participle dictionary.

7. according to the method described in any one in claim 1 to 6, wherein, the method is further comprising the steps of:

X, according to the content of other personages' that are associated with a personage people tag and/or described other personage's issues, upgrades described personage's people tag.

8. method according to claim 7, wherein, described step x comprises the following steps:

-according to other personages' associated with described figure picture address feature and network social intercourse relationship characteristic, other personages associated with described figure picture are divided into multiple groupings;

-according to the content of described other personages' that in described multiple groupings, all or part of grouping comprises people tag and/or described other personage's issues, obtain one or more people tag;

-according to described one or more people tag, upgrade described personage's people tag.

9. according to the method described in any one in claim 1 to 8, wherein, the method is further comprising the steps of:

-according to the content of described personage's issue, determine at least one people tag;

-according to described at least one people tag, upgrade described personage's people tag.

10. according to the method described in any one in claim 1 to 9, wherein, the method is further comprising the steps of:

-according to following at least one factor, determine the weight information of personage's people tag;

-the degree of concern of other personages to this personage with this people tag;

-this personage is to having other personages' the degree of concern of this people tag;

The quantity of the content of-this personage feature issue, that there is this people tag;

The degree that the content of-this personage feature issue, that have this people tag is paid close attention to by other personages;

The degree of concern of the content of-this personage to other personages feature issue, that there is this people tag;

Whether-this personage has the authentication information corresponding with this people tag.

11. 1 kinds for searching for personage's searcher, and wherein, this searcher comprises:

The first acquisition device, for obtaining the search sequence from user;

12. searchers according to claim 11, wherein, described the first determining device comprises:

List acquisition device, for obtaining personage's list of mating with each search participle of described at least one search participle;

Personage's determining device, for determining the personage who all exists in personage's list of all couplings acquisitions, as described at least one candidate personage.

13. according to the searcher described in claim 11 or 12, and wherein, described collator comprises:

The first weight determining device, for each the candidate personage to described at least one candidate personage, according to the weight information of people tag this candidate personage, that mate with described at least one search participle, determine the weight information that this candidate personage is corresponding with described search sequence;

The first sub-collator, for according to each candidate personage's weight information, sorts to all or part of candidate personage in described at least one candidate personage.

14. according to the searcher described in claim 11 or 12, and wherein, described collator comprises:

The second sub-collator, be used for according to the weight information of people tag each candidate personage, that mate with described search participle, and in conjunction with authentication information corresponding to each candidate personage, all or part of candidate personage in described at least one candidate personage is sorted.

15. according to claim 11 to the searcher described in any one in 14, and wherein, described the second acquisition device comprises:

Participle device, for described search sequence is carried out to participle, obtains participle list;

Selecting arrangement, for selecting important participle from described participle list, as described at least one search participle.

16. according to claim 11 to the searcher described in any one in 14, and wherein, described the second acquisition device comprises with lower device:

Sub-acquisition device: for described search sequence is carried out to participle, obtain described at least one search participle;

17. according to claim 11 to the searcher described in any one in 16, and wherein, this searcher also comprises:

The first updating device, for according to the content of other personages' that are associated with a personage people tag and/or described other personage's issues, upgrades described personage's people tag.

18. searchers according to claim 17, wherein, described the first updating device comprises:

Apparatus for grouping, for according to other personages' associated with described figure picture address feature and network social intercourse relationship characteristic, is divided into multiple groupings by other personages associated with described figure picture;

The first label determining device, for described other personages' the people tag and/or the content of described other personage's issues that comprise according to all or part of grouping of described multiple groupings, obtains one or more people tag;

Sub-updating device, for according to described one or more people tag, upgrades described personage's people tag.

19. according to claim 11 to the searcher described in any one in 18, and wherein, this searcher also comprises:

The second label determining device, for according to the content of described personage's issue, determines at least one people tag;

The second updating device, for according to described at least one people tag, upgrades described personage's people tag.

20. according to claim 11 to the searcher described in any one in 19, and wherein, this searcher also comprises:

The second weight determining device, for according to following at least one factor, determines the weight information of personage's people tag;

21. 1 kinds of computer equipments, comprise the searcher as described in claim 11 to 20 at least one.