CN103631970A - Method and device for mining associated relationship between attributes and entities - Google Patents
Method and device for mining associated relationship between attributes and entities Download PDFInfo
- Publication number
- CN103631970A CN103631970A CN201310714291.6A CN201310714291A CN103631970A CN 103631970 A CN103631970 A CN 103631970A CN 201310714291 A CN201310714291 A CN 201310714291A CN 103631970 A CN103631970 A CN 103631970A
- Authority
- CN
- China
- Prior art keywords
- entity
- attribute
- fructification
- entities
- sample cluster
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/288—Entity relationship models
Abstract
The invention provides a method and a device for mining an associated relationship between attributes and entities. The method comprises the following steps: obtaining attributes to be associated; obtaining at least one seed entity from multiple entities according to the attributes to be associated; and obtaining associated entities of the at least one seed entity, and associating the attributes to be associated with the at least one seed entity and the associated entities of the at least one seed entity. The method provided by the embodiment of the invention can be used for mining the multiple associated entities of the attributes to be associated, and realizing the mining of attributes specified by a user corresponding to the entities (namely, the attributes to be associated) in the same way, thus providing more comprehensive and more delicate detailed services with higher quality; the method also can be used for mining the associated relationship between the entities and the attributes specified by the user (namely, the attributes to be associated) in any field without being limited by application fields, so that the method is wide in application.
Description
Technical field
The present invention relates to field of computer technology, relate in particular to a kind of method and apparatus that excavates attribute and entity associated relation.
Background technology
Along with Internet technology, the particularly fast development of wireless interconnected network technology, it is more and more general that information service becomes.When information service provider provides information service, for example, search engine provides search service etc., conventionally can excavate the incidence relation between entity and attribute, and provides information service according to the incidence relation between entity and attribute.Particularly, the objective things in real world can be called to entity, such as concept, things or event etc.For instance, company of ”, Baidu of movie and television play “Wo Shi special technical soldier and Big Bang Theory are all the examples of entity.Meanwhile, each entity has attribute, the relevant information of attribute reflection entity, and for example, army's subject matter, company office, modern universe theory are respectively the attributes that above-mentioned entity is corresponding.
The method of obtaining at present incidence relation between entity and attribute is mainly directedly from the structural data of website to capture entity attribute pair, and according to entity attribute to setting up the incidence relation between entity and attribute.But, mainly there is following problem, because an attribute corresponding to entity is diversified, a corresponding entity, the attribute obtaining from website is some aspects, and this attribute possibly cannot well meet user's demand.Therefore prior art cannot be excavated the corresponding user-specific attributes of entity, for example, cannot excavate certain film and belong to " counteroffensive of Cock silk " attribute etc., similarly, also cannot excavate the entities corresponding to attribute such as " counteroffensive of Cock silk ", " curing system ", " the cruel heart ", as the film of " counteroffensive of Cock silk " subject matter, novel etc.
Summary of the invention
The present invention is intended at least one of solve the problems of the technologies described above.
For this reason, first object of the present invention is to propose a kind of method of excavating attribute and entity associated relation.The method can be excavated a plurality of associated entity of attribute to be associated, in like manner realize to excavate the corresponding user-specific attributes of entity (being attribute to be associated), thereby provides more comprehensively, detailed service meticulousr, more high-quality.
Second object of the present invention is to propose a kind of device that excavates attribute and entity associated relation.
To achieve these goals, the excavation attribute of first aspect present invention embodiment and the method for entity associated relation, comprise the following steps: obtain attribute to be associated; According to described attribute to be associated, from a plurality of entities, obtain at least one and plant fructification; And the associated entity that obtains described at least one kind fructification, and described attribute to be associated is associated with the associated entity of described at least one kind fructification, described at least one kind fructification.
The excavation attribute of the embodiment of the present invention and the method for entity associated relation, by attribute to be associated, obtain kind of a fructification, according to kind of fructification, obtain relevant associated entity again, thus, can excavate a plurality of associated entity of attribute to be associated, in like manner realize excavate the corresponding user-specific attributes of entity (being attribute to be associated), thereby provide more comprehensively, detailed service meticulousr, more high-quality, for example, according to user-specific attributes to user's recommended entity; According to the method, can also excavate the incidence relation between any domain entities and given attribute (being attribute to be associated), not be subject to the restriction of application, be widely used.
To achieve these goals, the excavation attribute of second aspect present invention embodiment and the device of entity associated relation, comprising: attribute acquisition module to be associated, for obtaining attribute to be associated; Plant fructification acquisition module, for obtaining at least one according to described attribute to be associated from a plurality of entities, plant fructification; Associated entity acquisition module, for obtaining the associated entity of described at least one kind fructification; And relating module, for described attribute to be associated is associated with the associated entity of described at least one kind fructification, described at least one kind fructification.
The excavation attribute of the embodiment of the present invention and the device of entity associated relation, by attribute acquisition module to be associated, obtain attribute to be associated, then plant fructification acquisition module and obtain kind of a fructification according to attribute to be associated, associated entity acquisition module obtains the associated entity of kind of fructification according to kind of fructification afterwards, thus, can excavate a plurality of associated entity of attribute to be associated, in like manner realize and excavate the corresponding user-specific attributes of entity (being attribute to be associated), thereby provide more comprehensively, meticulousr, the more detailed service of high-quality, for example, according to user-specific attributes to user's recommended entity, according to this device, can also excavate the incidence relation between any domain entities and user-specific attributes (being attribute to be associated), not be subject to the restriction of application, be widely used.
The aspect that the present invention is additional and advantage in the following description part provide, and part will become obviously from the following description, or recognize by practice of the present invention.
Accompanying drawing explanation
Above-mentioned and/or the additional aspect of the present invention and advantage will become from the following description of the accompanying drawings of embodiments and obviously and easily understand, wherein,
Fig. 1 is the process flow diagram that excavates according to an embodiment of the invention the method for attribute and entity associated relation;
Fig. 2 is the process flow diagram that excavates according to an embodiment of the invention the method for attribute and entity associated relation;
Fig. 3 is the process flow diagram that obtains according to an embodiment of the invention distributional difference value;
Fig. 4 obtains the process flow diagram that obtains associated entity according to an embodiment of the invention;
Fig. 5 is the structural representation that excavates according to an embodiment of the invention the device of attribute and entity associated relation;
Fig. 6 is the structural representation that excavates according to an embodiment of the invention the device of attribute and entity associated relation.
Embodiment
Describe embodiments of the invention below in detail, the example of described embodiment is shown in the drawings, and wherein same or similar label represents same or similar element or has the element of identical or similar functions from start to finish.Below by the embodiment being described with reference to the drawings, be exemplary, only for explaining the present invention, and can not be interpreted as limitation of the present invention.On the contrary, embodiments of the invention comprise spirit and all changes within the scope of intension, modification and the equivalent that falls into additional claims.
In description of the invention, it will be appreciated that, term " first ", " second " etc. are only for describing object, and can not be interpreted as indication or hint relative importance.In description of the invention, it should be noted that, unless otherwise clearly defined and limited, term " is connected ", " connection " should be interpreted broadly, and for example, can be to be fixedly connected with, and can be also to removably connect, or connects integratedly; Can be mechanical connection, can be to be also electrically connected to; Can be to be directly connected, also can indirectly be connected by intermediary.For the ordinary skill in the art, can concrete condition understand above-mentioned term concrete meaning in the present invention.In addition,, in description of the invention, except as otherwise noted, the implication of " a plurality of " is two or more.
In process flow diagram or any process of otherwise describing at this or method describe and can be understood to, represent to comprise that one or more is for realizing module, fragment or the part of code of executable instruction of the step of specific logical function or process, and the scope of the preferred embodiment of the present invention comprises other realization, wherein can be not according to order shown or that discuss, comprise according to related function by the mode of basic while or by contrary order, carry out function, this should be understood by embodiments of the invention person of ordinary skill in the field.
In order to excavate the incidence relation between entity and user-specific attributes (as user-specific attributes) in any field, thereby to user provide more comprehensively, meticulousr information service, the present invention proposes a kind of method and apparatus that excavates attribute and entity associated relation.Below with reference to accompanying drawing, the excavation attribute of the embodiment of the present invention and the method and apparatus of entity associated relation are described.
A method of excavating attribute and entity associated relation, comprises the following steps: obtain attribute to be associated; According to attribute to be associated, from a plurality of entities, obtain at least one and plant fructification; And obtain the associated entity that at least one plants fructification, and the associated entity that attribute to be associated and at least one are planted fructification, at least one kind fructification is associated.
Fig. 1 is the process flow diagram that excavates according to an embodiment of the invention the method for attribute and entity associated relation.
As shown in Figure 1, the method for excavation attribute and entity associated relation comprises the steps.
Step S101, obtains attribute to be associated.
In one embodiment of the invention, attribute to be associated is the attribute of the features such as a class description user impression, product performance.Attribute to be associated can, with netspeak real-time update, for example, can obtain attribute to be associated to a plurality of webpage analyses.For instance, can there be " counteroffensive of Cock silk ", " evilness is defeated justice ", " curing system ", " the cruel heart ", " exposing the wealth " etc. to describe the attribute to be associated of user's impression; For product entity, can there be " cost performance is high ", " durable " etc. to describe the attribute to be associated of user's experience.
Step S102 obtains at least one and plants fructification from a plurality of entities according to attribute to be associated.
Particularly, after obtaining attribute to be associated, according to attribute to be associated, from a plurality of entities, obtain at least one and plant fructification.Wherein, using the entity name tight with attribute relationship to be associated, the degree of correlation is high as planting fructification.For example, if attribute to be associated is " curing system ", the kind fructification of obtaining can be that the movie and television play entity of " curing system " is, the caricature entity of the novel entity of " curing system ", " curing system " or other entity of " curing system " etc.This process is relevant with the degree of association of entity with the degree of association, the service application of user and entity, in subsequent embodiment, will describe in detail.
Step S103, obtains the associated entity that at least one plants fructification, and the associated entity that attribute to be associated and at least one are planted fructification, at least one kind fructification is associated.
Particularly, from a plurality of entities, obtain at least one and plant after fructification, then by centered by least one kind fructification, obtain at least one and plant the higher associated entity of the fructification degree of correlation.Take that from a plurality of entities, to have obtained a kind fructification be example, for example, if the kind fructification obtaining from a plurality of entities is the movie and television play seed entity A of " curing system ", then obtaining the associated entity of the movie and television play seed entity A that should " cure system ", can be that the novel entity B of " curing system " is, other entities E of the caricature entity C of " curing system ", " curing system " or other movie and television play seed F that " cures system " and G etc. such as the associated entity obtaining.This process can expand the scope of entity, recalls some associated entity.
More specifically, after obtaining at least one associated entity of planting fructification, the associated entity of attribute to be associated and at least one being planted to fructification, at least one kind fructification is associated.For example, after obtaining the novel entity or other movie and television play entities of " curing system " that associated entity " cures system ", attribute to be associated " is cured to system " with the movie and television play kind fructification of " curing system ", the associated entity of the movie and television play kind fructification of " curing system " (the movie and television play entities of the novel entity of " curing system " or other " healing is ") is associated.
Wherein, the operation being associated can be to attribute to be associated, that at least one plants fructification, at least one plants the associated entity of fructification is labelled or set up corresponding relation between them etc.For example, can be by the movie and television play kind fructification of attribute to be associated " cure system " and " curing system ", the associated entity of the movie and television play kind fructification of " curing system " (the novel entity of " curing system " or other movie and television play entities of " curing system ") stick the label of " healing is " or set up corresponding relation between them etc.
The excavation attribute of the embodiment of the present invention and the method for entity associated relation, by attribute to be associated, obtain kind of a fructification, according to kind of fructification, obtain relevant associated entity again, thus, can excavate a plurality of associated entity of attribute to be associated, in like manner realize excavate the corresponding user-specific attributes of entity (being attribute to be associated), thereby provide more comprehensively, detailed service meticulousr, more high-quality, for example, according to user-specific attributes to user's recommended entity; According to the method, can also excavate the incidence relation between any domain entities and given attribute (being attribute to be associated), not be subject to the restriction of application, be widely used.
Fig. 2 is the process flow diagram that excavates in accordance with another embodiment of the present invention the method for attribute and entity associated relation.In an embodiment of the present invention, adopt the mode of distributional difference from a plurality of entities, to obtain kind of a fructification.
Particularly, as shown in Figure 2, the method for excavating attribute and entity associated relation comprises the steps.
Step S201, obtains attribute to be associated.
In one embodiment of the invention, attribute to be associated is the attribute of the features such as a class description user impression, product performance.Attribute to be associated can, with netspeak real-time update, for example, can obtain attribute to be associated to a plurality of webpage analyses.For instance, can there be " counteroffensive of Cock silk ", " evilness is defeated justice ", " curing system ", " the cruel heart ", " exposing the wealth " etc. to describe the attribute to be associated of user's impression; For product entity, can there be " cost performance is high ", " durable " etc. to describe the attribute to be associated of user's experience.
Step S202 obtains a plurality of entities from default entity storehouse.
Particularly, the entity storehouse of default entity storehouse for obtaining from network in advance, default entity stores a plurality of entities in storehouse, and wherein, default entity storehouse can be stored in server or in miscellaneous equipment.Can also classify to default entity storehouse, different application services can have different default entity storehouses.
Step S203 obtains the associated user sample cluster with attribute to be associated from overall user sample cluster.
Particularly, according to attribute to be associated, from overall user sample cluster, obtain the associated user sample cluster with attribute to be associated.For example, if attribute to be associated is " exposing the wealth ", overall user sample cluster is 1,000 ten thousand and watches the user of movie and television play, obtains so 1,000,000 users that watch " exposing the wealth " movie and television play in overall user sample cluster, has the associated user sample cluster of attribute to be associated.
Step S204, obtains respectively a plurality of distributional difference values of a plurality of entities in associated user sample cluster.
Particularly, same entity is different at overall user sample cluster and the distribution that has in the associated user sample cluster of attribute to be associated.The size of distributional difference value can be corresponding the height of the degree of correlation that embodies entity and attribute to be associated, be convenient to follow-uply according to distributional difference value, entity be screened.Obtaining of distributional difference value will describe in detail in subsequent embodiment particularly.
Step S205, screens to obtain at least one according to a plurality of distributional difference values to a plurality of entities and plants fructification.
Particularly, obtain after a plurality of distributional difference values of a plurality of entities in associated user sample cluster, according to a plurality of distributional difference values, a plurality of entities are screened to obtain at least one and plant fructification.Wherein, plant with the to be associated attributes correlation higher entity of fructification for screening from a plurality of entities according to distributional difference value.
Step S206, obtains the associated entity that at least one plants fructification, and the associated entity that attribute to be associated and at least one are planted fructification, at least one kind fructification is associated.
Particularly, from a plurality of entities, obtain at least one and plant after fructification, then by centered by least one kind fructification, obtain at least one and plant the higher associated entity of the fructification degree of correlation.Take that from a plurality of entities, to have obtained a kind fructification be example, for example, if the kind fructification obtaining from a plurality of entities is the movie and television play seed entity A of " curing system ", then obtaining the associated entity of the movie and television play seed entity A that should " cure system ", can be that the novel entity B of " curing system " is, other entities E of the caricature entity C of " curing system ", " curing system " or other movie and television play seed F that " cures system " and G etc. such as the associated entity obtaining.This process can expand the scope of entity, recalls some associated entity.
More specifically, after obtaining at least one associated entity of planting fructification, the associated entity of attribute to be associated and at least one being planted to fructification, at least one kind fructification is associated.For example, after obtaining the novel entity or other movie and television play entities of " curing system " that associated entity " cures system ", attribute to be associated " is cured to system " with the movie and television play kind fructification of " curing system ", the associated entity of the movie and television play kind fructification of " curing system " (the movie and television play entities of the novel entity of " curing system " or other " healing is ") is associated.
Wherein, the operation being associated can be to attribute to be associated, that at least one plants fructification, at least one plants the associated entity of fructification is labelled or set up corresponding relation between them etc.For example, can be by the movie and television play kind fructification of attribute to be associated " cure system " and " curing system ", the associated entity of the movie and television play kind fructification of " curing system " (the novel entity of " curing system " or other movie and television play entities of " curing system ") stick the label of " healing is " or set up corresponding relation between them etc.
The excavation attribute of the embodiment of the present invention and the method for entity associated relation, adopt distributional difference value from a plurality of entities, to obtain kind of a fructification, distributional difference value reflects the distribution of kind of fructification truly, the kind fructification of obtaining and the degree of correlation of attribute to be associated are higher, more accurate, thereby further promote the quality of information service.
Fig. 3 is the process flow diagram that obtains according to an embodiment of the invention distributional difference value.In one embodiment of the invention, as shown in Figure 3, step S204 specifically comprises:
S2041, obtains respectively a plurality of first distribution proportions of a plurality of users relevant to a plurality of entities in overall user sample cluster.
For example, overall user sample cluster is 1,000 ten thousand and watches the user of movie and television play, wherein have 500,000 user to watch movie and television play entity M, the distribution proportion of the user who watches so movie and television play entity M in overall user sample cluster is 500,000 divided by 1,000 ten thousand, and the first distribution proportion is 5%.Similarly, obtain successively a plurality of first distribution proportions of a plurality of users relevant to a plurality of entities in overall user sample cluster.
S2042, obtains respectively the second distribution proportion of a plurality of users relevant to a plurality of entities in associated user sample cluster.
For example, attribute to be associated is " exposing the wealth ", associated user sample cluster is 1,000,000 users that watch " exposing the wealth " movie and television play, wherein, 300000 users have watched movie and television play entity M, the distribution proportion of the user who watches so movie and television play entity M in associated user sample cluster is 300,000 divided by 1,000,000, and the second distribution proportion is 30%.Similarly, obtain successively a plurality of second distribution proportions of a plurality of users relevant to a plurality of entities in associated user sample cluster.
S2043, obtains distributional difference value according to the second distribution proportion and the first distribution proportion.
Particularly, according to the second distribution proportion obtaining and the first distribution proportion, with the second distribution proportion, divided by the first distribution proportion, obtain distributional difference value.
For example, overall user sample cluster is 1,000 ten thousand and watches the user of movie and television play wherein have 500,000 user to watch movie and television play entity M, and the first distribution proportion is 5% so; If attribute to be associated is " exposing the wealth ", associated user sample cluster is 1,000,000 users that watch " exposing the wealth " movie and television play, wherein, 300000 users have watched movie and television play entity M, the second distribution proportion is 30% so, uses 30% divided by 5%, and obtaining distributional difference value is 6.Wherein distributional difference value is larger, illustrates that the degree of correlation that movie and television play entity M and attribute to be associated " expose the wealth " is higher.
Thus, the distributional difference value of obtaining according to the first distribution proportion and the second distribution proportion more can embody the degree of association, and distributional difference value is more accurate.
In one embodiment of the invention, in step S205, overall user sample cluster is a plurality of, corresponding a plurality of network application services respectively, the distributional difference value that each entity is corresponding is a plurality of, according to a plurality of distributional difference values, a plurality of entities is screened to obtain described at least one kind fructification (being step S205) and also comprises: according to default distributional difference value screening rule, described a plurality of entities are screened; Or, create distributional difference value sorter, and according to distributional difference value sorter, a plurality of entities are screened, in addition, can also use other method.
Particularly, take that entity is known in associated user sample cluster, Baidu's mhkc, Baidu below, the distributional difference in Baidu's session illustrates the method for a plurality of entities being screened according to default distributional difference value screening rule as example.The screening rule that the method adopts is as follows:
(1) output entity in associated user sample cluster, Baidu's mhkc, Baidu, know, the larger entity of distributional difference value in Baidu's session, with Suser, Stieba, Siknow, Ssession respectively presentation-entity in associated user sample cluster, Baidu's mhkc, Baidu, know, distributional difference value in Baidu's session, as the entity of output Suser>10, Stieba>50, Siknow>50 or Ssession>30;
(2) in output Stieba, Siknow, Ssession, have at least one be greater than 3 and Suser be also greater than 3 entity;
(3) output Stieba, Siknow, Ssession are all greater than 3 entity;
(4) in output Stieba, Siknow, Ssession, have at least one to be greater than 3, one and to be greater than 8 entity.
Can also set up sorter according to above-mentioned screening rule, for example, can adopt the method for setting up sorter of prior art to set up classification, the foundation of sorter can be raised the efficiency.The foundation of sorter can adopt prior art, does not repeat them here.
Above-mentioned at least one method accuracy rate of planting fructification of screening in a plurality of entities according to distributional difference value is high, but the entity below threshold value can not be called back in the screening rule of setting, for this reason the follow-up associated entity that also needs to obtain kind of fructification.
Fig. 4 obtains the process flow diagram that obtains associated entity according to an embodiment of the invention.In one embodiment of the invention, as shown in Figure 4, in step S206, obtain at least one associated entity of planting fructification and specifically comprise:
S2061, obtains respectively at least one and plants fructification to the first incidence relation having between user's sample cluster of attribute to be associated.
Particularly, for example, can kind of fructification be described for example, to the first incidence relation having between user's sample cluster of attribute to be associated, matrix A by matrix.
S2062, obtains the associated entity group of user's sample cluster with attribute to be associated, and obtain there is attribute to be associated user's sample cluster to the second incidence relation between associated entity group.
Particularly, obtain the associated entity group of user's sample cluster with attribute to be associated, for example, if there is user's sample cluster of attribute to be associated for watching the user of the movie and television play entity of " curing system ", obtain the movie and television play entity, " curing system " novel entity, " curing system " caricature entity of " cure system " or other entity of " curing system ", be the associated entity group of user's sample cluster with attribute to be associated.
More specifically, can by matrix describe there is attribute to be associated user's sample cluster for example, to the second incidence relation between associated entity group, matrix B.
S2063, obtains respectively at least one kind fructification to associated entity group's the 3rd incidence relation according to the first incidence relation and the second incidence relation.
Particularly, for example, can obtain at least one kind fructification to associated entity group's the 3rd incidence relation according to matrix A and matrix B, can describe by Matrix C.For example, can get Matrix C by simple matrix multiple, can also be weighted to process and multiply each other again afterwards.
S2064, screens to obtain to each associated entity in associated entity group the associated entity that at least one plants fructification according to the 3rd incidence relation.
For example, the 3rd incidence relation can identify by Matrix C, each element in Matrix C is that this entity seed is to the degree of correlation information between associated entity, according to this matrix, can obtain kind of fructification to the similarity of paths pathsim feature on the path of each associated entity, according to this feature, obtain the associated entity of kind of fructification.In addition, pathsim feature can also find and be equal to entity peer objects, reduces the impact of popular entity.Wherein, the computing formula of Pathsim feature is as follows:
Wherein, a
ibe i entity, a
jbe j entity, pc
r(a
i, a
j) be that in Matrix C, i element value capable, j row (is entity a
iwith entity a
jbetween the degree of correlation), pc
r(a
i, a
i) be that in Matrix C, i element value capable, i row (is entity a
ithe degree of correlation of self), pc
r-1(a
j, a
i) be the inverse matrix C of Matrix C
-1in the element value of capable, the i of j row, pc
r-1(a
j, a
j) be the inverse matrix C of Matrix C
-1in the element value of capable, the j of j row.
Filter the entity that above-mentioned association of obtaining.Particularly, can setting threshold filter out doubtful incoherent entity in the entity that association goes out, wherein, threshold value can be the multiple of the distributional difference value of seed entity on associated user sample cluster, for example 2 times, 3 times or other multiple.
Thus, the 3rd incidence relation obtaining has more directly reacted the associated entity of planting fructification, makes the associated entity of acquisition more accurate.
In order to realize above-described embodiment, the present invention also proposes a kind of device that excavates attribute and entity associated relation.
A device that excavates attribute and entity associated relation, comprising: attribute acquisition module to be associated, for obtaining attribute to be associated; Plant fructification acquisition module, for obtaining at least one according to attribute to be associated from a plurality of entities, plant fructification; Associated entity acquisition module, the associated entity of planting fructification for obtaining at least one; And relating module, for attribute to be associated and at least one being planted to fructification, associated entity that at least one plants fructification is associated.
Fig. 5 is the structural representation that excavates according to an embodiment of the invention the device of attribute and entity associated relation.
As shown in Figure 5, the device of excavation attribute and entity associated relation comprises: attribute acquisition module 100 to be associated, kind fructification acquisition module 200, associated entity acquisition module 300 and relating module 400.
Wherein, attribute acquisition module 100 to be associated is for obtaining attribute to be associated.
Particularly, attribute to be associated is the attribute of the features such as a class description user impression, product performance.Attribute to be associated can, with netspeak real-time update, for example, can obtain attribute to be associated to a plurality of webpage analyses.For instance, can there be " counteroffensive of Cock silk ", " evilness is defeated justice ", " curing system ", " the cruel heart ", " exposing the wealth " etc. to describe the attribute to be associated of user's impression; For product entity, can there be " cost performance is high ", " durable " etc. to describe the attribute to be associated of user's experience.
Plant fructification acquisition module 200 and plant fructification for obtaining at least one according to attribute to be associated from a plurality of entities.
Particularly, after obtaining attribute to be associated, according to attribute to be associated, from a plurality of entities, obtain at least one and plant fructification.Wherein, using the entity name tight with attribute relationship to be associated, the degree of correlation is high as planting fructification.For example, if attribute to be associated is " curing system ", the kind fructification of obtaining can be that the movie and television play entity of " curing system " is, the caricature entity of the novel entity of " curing system ", " curing system " or other entity of " curing system " etc.This process is relevant with the degree of association of entity with the degree of association, the service application of user and entity, in subsequent embodiment, will describe in detail.
The associated entity that associated entity acquisition module 300 is planted fructification for obtaining at least one.
Particularly, from a plurality of entities, obtain at least one and plant after fructification, then by centered by least one kind fructification, obtain at least one and plant the higher associated entity of the fructification degree of correlation.Take that from a plurality of entities, to have obtained a kind fructification be example, for example, if the kind fructification obtaining from a plurality of entities is the movie and television play seed entity A of " curing system ", then obtaining the associated entity of the movie and television play seed entity A that should " cure system ", can be that the novel entity B of " curing system " is, other entities E of the caricature entity C of " curing system ", " curing system " or other movie and television play seed F that " cures system " and G etc. such as the associated entity obtaining.This process can expand the scope of entity, recalls some associated entity.
Relating module 400 for attribute to be associated and at least one being planted to fructification, associated entity that at least one plants fructification is associated.
Particularly, after obtaining at least one associated entity of planting fructification, the associated entity of attribute to be associated and at least one being planted to fructification, at least one kind fructification is associated.
For example, after obtaining the novel entity or other movie and television play entities of " curing system " that associated entity " cures system ", attribute to be associated " is cured to system " with the movie and television play kind fructification of " curing system ", the associated entity of the movie and television play kind fructification of " curing system " (the movie and television play entities of the novel entity of " curing system " or other " healing is ") is associated.
Wherein, the operation being associated can be to attribute to be associated, that at least one plants fructification, at least one plants the associated entity of fructification is labelled or set up corresponding relation between them etc.For example, can be by the movie and television play kind fructification of attribute to be associated " cure system " and " curing system ", the associated entity of the movie and television play kind fructification of " curing system " (the novel entity of " curing system " or other movie and television play entities of " curing system ") stick the label of " healing is " or set up corresponding relation between them etc.
The excavation attribute of the embodiment of the present invention and the device of entity associated relation, by attribute acquisition module to be associated, obtain attribute to be associated, then plant fructification acquisition module and obtain kind of a fructification according to attribute to be associated, associated entity acquisition module obtains the associated entity of kind of fructification according to kind of fructification afterwards, thus, can excavate a plurality of associated entity of attribute to be associated, in like manner realize and excavate the corresponding user-specific attributes of entity (being attribute to be associated), thereby provide more comprehensively, meticulousr, the more detailed service of high-quality, for example, according to user-specific attributes to user's recommended entity, according to this device, can also excavate the incidence relation between any domain entities and user-specific attributes (being attribute to be associated), not be subject to the restriction of application, be widely used.
Fig. 6 is the structural representation that excavates according to an embodiment of the invention the device of attribute and entity associated relation.
As shown in Figure 6, the device of excavation attribute and entity associated relation comprises: attribute acquisition module 100 to be associated, kind fructification acquisition module 200, entity acquiring unit 210, associated user sample cluster acquiring unit 220, distributional difference value acquiring unit 230, screening unit 240, associated entity acquisition module 300, the first incidence relation acquiring unit 310, the second incidence relation acquiring unit 320, the 3rd incidence relation acquiring unit 330, screening unit 340 and relating module 400.Wherein, plant fructification acquisition module 200 and comprise entity acquiring unit 210, associated user sample cluster acquiring unit 220, distributional difference value acquiring unit 230, screening unit 240; Associated entity acquisition module 300 comprises the first incidence relation acquiring unit 310, the second incidence relation acquiring unit 320, the 3rd incidence relation acquiring unit 330, screening unit 340.
In one embodiment of the invention, the first incidence relation acquiring unit 310, the second incidence relation acquiring unit 320, the 3rd incidence relation acquiring unit 330, screening unit 340 are optional.
Particularly, attribute acquisition module 100 to be associated is for obtaining attribute to be associated.
In one embodiment of the invention, attribute to be associated is the attribute of the features such as a class description user impression, product performance.Attribute to be associated can, with netspeak real-time update, for example, can obtain attribute to be associated to a plurality of webpage analyses.For instance, can there be " counteroffensive of Cock silk ", " evilness is defeated justice ", " curing system ", " the cruel heart ", " exposing the wealth " etc. to describe the attribute to be associated of user's impression; For product entity, can there be " cost performance is high ", " durable " etc. to describe the attribute to be associated of user's experience.
Entity acquiring unit 210 is for obtaining a plurality of entities from default entity storehouse.
Particularly, the entity storehouse of default entity storehouse for obtaining from network in advance, default entity stores a plurality of entities in storehouse, and wherein, default entity storehouse can be stored in server or in miscellaneous equipment.Can also classify to default entity storehouse, different application services can have different default entity storehouses.
Associated user sample cluster acquiring unit 220 is for obtaining the associated user sample cluster with attribute to be associated from overall user sample cluster.
Particularly, according to attribute to be associated, from overall user sample cluster, obtain the associated user sample cluster with attribute to be associated.For example, if attribute to be associated is " exposing the wealth ", overall user sample cluster is 1,000 ten thousand and watches the user of movie and television play, obtains so 1,000,000 users that watch " exposing the wealth " movie and television play in overall user sample cluster, has the associated user sample cluster of attribute to be associated.
Distributional difference value acquiring unit 230 is for obtaining respectively a plurality of entities in a plurality of distributional difference values of associated user sample cluster.
Particularly, same entity is different at overall user sample cluster and the distribution that has in the associated user sample cluster of attribute to be associated.The size of distributional difference value can be corresponding the height of the degree of correlation that embodies entity and attribute to be associated, be convenient to follow-uply according to distributional difference value, entity be screened.Obtaining of distributional difference value will describe in detail in subsequent embodiment particularly.
In one embodiment of the invention, distributional difference value acquiring unit 230 also specifically for: obtain respectively a plurality of first distribution proportions of a plurality of users relevant to a plurality of entities in overall user sample cluster, and obtain respectively the second distribution proportion of a plurality of users relevant to a plurality of entities in associated user sample cluster, and obtain distributional difference value according to the second distribution proportion and the first distribution proportion.
Wherein, illustrate obtaining of the first distribution proportion below, for example, overall user sample cluster is 1,000 ten thousand and watches the user of movie and television play, wherein there is 500,000 user to watch movie and television play entity M, the distribution proportion of the user who watches so movie and television play entity M in overall user sample cluster is 500,000 divided by 1,000 ten thousand, and the first distribution proportion is 5%.Similarly, obtain successively a plurality of first distribution proportions of a plurality of users relevant to a plurality of entities in overall user sample cluster.
Illustrate obtaining of the second distribution proportion below, for example, attribute to be associated is " exposing the wealth ", associated user sample cluster is 1,000,000 users that watch " exposing the wealth " movie and television play, wherein, 300000 users have watched movie and television play entity M, and the distribution proportion of the user who watches so movie and television play entity M in associated user sample cluster is 300,000 divided by 1,000,000, and the second distribution proportion is 30%.Similarly, obtain successively a plurality of second distribution proportions of a plurality of users relevant to a plurality of entities in associated user sample cluster.
According to the second distribution proportion obtaining and the first distribution proportion, with the second distribution proportion, divided by the first distribution proportion, obtain distributional difference value.For example, overall user sample cluster is 1,000 ten thousand and watches the user of movie and television play wherein have 500,000 user to watch movie and television play entity M, and the first distribution proportion is 5% so; If attribute to be associated is " exposing the wealth ", associated user sample cluster is 1,000,000 users that watch " exposing the wealth " movie and television play, wherein, 300000 users have watched movie and television play entity M, the second distribution proportion is 30% so, uses 30% divided by 5%, and obtaining distributional difference value is 6.Wherein distributional difference value is larger, illustrates that the degree of correlation that movie and television play entity M and attribute to be associated " expose the wealth " is higher.
Thus, the distributional difference value of obtaining according to the first distribution proportion and the second distribution proportion more can embody the degree of association, and distributional difference value is more accurate.
Fructification is planted for a plurality of entities being screened to obtain at least one according to a plurality of distributional difference values in screening unit 240.
Particularly, obtain after a plurality of distributional difference values of a plurality of entities in associated user sample cluster, according to a plurality of distributional difference values, a plurality of entities are screened to obtain at least one and plant fructification.Wherein, plant with the to be associated attributes correlation higher entity of fructification for screening from a plurality of entities according to distributional difference value.
In addition, overall user sample cluster is a plurality of, corresponding a plurality of network application services respectively, the distributional difference value that each entity is corresponding is a plurality of, and screening unit 240 also screens and comprises a plurality of entities according to a plurality of distributional difference values: according to default distributional difference value screening rule, a plurality of entities are screened; Or, create distributional difference value sorter, and according to distributional difference value sorter, a plurality of entities are screened, in addition, can also use other method.
Particularly, take that entity is known in associated user sample cluster, Baidu's mhkc, Baidu below, the distributional difference in Baidu's session illustrates the method for a plurality of entities being screened according to default distributional difference value screening rule as example.The screening rule that the method adopts is as follows:
(1) output entity in associated user sample cluster, Baidu's mhkc, Baidu, know, the larger entity of distributional difference value in Baidu's session, with Suser, Stieba, Siknow, Ssession respectively presentation-entity in associated user sample cluster, Baidu's mhkc, Baidu, know, distributional difference value in Baidu's session, as the entity of output Suser>10, Stieba>50, Siknow>50 or Ssession>30;
(2) in output Stieba, Siknow, Ssession, have at least one be greater than 3 and Suser be also greater than 3 entity;
(3) output Stieba, Siknow, Ssession are all greater than 3 entity;
(4) in output Stieba, Siknow, Ssession, have at least one to be greater than 3, one and to be greater than 8 entity.
Can also set up sorter according to above-mentioned screening rule, for example, can adopt the method for setting up sorter of prior art to set up classification, the foundation of sorter can be raised the efficiency.The foundation of sorter can adopt prior art, does not repeat them here.
Above-mentioned at least one method accuracy rate of planting fructification of screening in a plurality of entities according to distributional difference value is high, but the entity below threshold value can not be called back in the screening rule of setting, for this reason the follow-up associated entity that also needs to obtain kind of fructification.
The first incidence relation acquiring unit 310 is planted fructification to the first incidence relation having between user's sample cluster of attribute to be associated for obtaining respectively at least one.
Particularly, for example, can kind of fructification be described for example, to the first incidence relation having between user's sample cluster of attribute to be associated, matrix A by matrix.
The second incidence relation acquiring unit 320 is for obtaining the associated entity group of user's sample cluster with attribute to be associated, and obtain there is attribute to be associated user's sample cluster to the second incidence relation between associated entity group.
Particularly, obtain the associated entity group of user's sample cluster with attribute to be associated, for example, if there is user's sample cluster of attribute to be associated for watching the user of the movie and television play entity of " curing system ", obtain the movie and television play entity, " curing system " novel entity, " curing system " caricature entity of " cure system " or other entity of " curing system ", be the associated entity group of user's sample cluster with attribute to be associated.
More specifically, can by matrix describe there is attribute to be associated user's sample cluster for example, to the second incidence relation between associated entity group, matrix B.
The 3rd incidence relation acquiring unit 330 is for obtaining respectively at least one kind fructification to associated entity group's the 3rd incidence relation according to the first incidence relation and the second incidence relation.
Particularly, for example, can obtain at least one kind fructification to associated entity group's the 3rd incidence relation according to matrix A and matrix B, can describe by Matrix C.For example, can get Matrix C by simple matrix multiple, can also be weighted to process and multiply each other again afterwards.
Screening unit 340 is for screening to obtain to associated entity described in each of associated entity group the associated entity that at least one plants fructification according to the 3rd incidence relation.
For example, the 3rd incidence relation can identify by Matrix C, each element in Matrix C is that this entity seed is to the degree of correlation information between associated entity, according to this matrix, can obtain kind of fructification to the similarity of paths pathsim feature on the path of each associated entity, according to this feature, obtain the associated entity of kind of fructification.In addition, pathsim feature can also find and be equal to entity peer objects, reduces the impact of popular entity.Wherein, the computing formula of Pathsim feature is as follows:
Wherein, a
ibe i entity, a
jbe j entity, pc
r(a
i, a
j) be that in Matrix C, i element value capable, j row (is entity a
iwith entity a
jbetween the degree of correlation), pc
r(a
i, a
i) be that in Matrix C, i element value capable, i row (is entity a
ithe degree of correlation of self), pc
r-1(a
j, a
i) be the inverse matrix C of Matrix C
-1in the element value of capable, the i of j row, pc
r-1(a
j, a
j) be the inverse matrix C of Matrix C
-1in the element value of capable, the j of j row.
Filter the entity that above-mentioned association of obtaining.Particularly, can setting threshold filter out doubtful incoherent entity in the entity that association goes out, wherein, threshold value can be the multiple of the distributional difference value of seed entity on associated user sample cluster, for example 2 times, 3 times or other multiple.
Thus, the 3rd incidence relation obtaining has more directly reacted the associated entity of planting fructification, makes the associated entity of acquisition more accurate.
Relating module 400 for attribute to be associated and at least one being planted to fructification, associated entity that at least one plants fructification is associated.
Particularly, after obtaining at least one associated entity of planting fructification, the associated entity of attribute to be associated and at least one being planted to fructification, at least one kind fructification is associated.
For example, after obtaining the novel entity or other movie and television play entities of " curing system " that associated entity " cures system ", attribute to be associated " is cured to system " with the movie and television play kind fructification of " curing system ", the associated entity of the movie and television play kind fructification of " curing system " (the movie and television play entities of the novel entity of " curing system " or other " healing is ") is associated.
Wherein, the operation being associated can be to attribute to be associated, that at least one plants fructification, at least one plants the associated entity of fructification is labelled or set up corresponding relation between them etc.For example, can be by the movie and television play kind fructification of attribute to be associated " cure system " and " curing system ", the associated entity of the movie and television play kind fructification of " curing system " (the novel entity of " curing system " or other movie and television play entities of " curing system ") stick the label of " healing is " or set up corresponding relation between them etc.Thus, the 3rd incidence relation obtaining has more directly reacted the associated entity of planting fructification, makes the associated entity of acquisition more accurate.
The excavation attribute of the embodiment of the present invention and the device of entity associated relation, the distributional difference value that distributional difference value acquiring unit obtains according to the first distribution proportion and the second distribution proportion more can embody the degree of association, and distributional difference value is more accurate; The 3rd incidence relation that the 3rd incidence relation acquiring unit obtains according to the first incidence relation and the second incidence relation has more directly reacted the associated entity of planting fructification, makes the associated entity that obtains more accurate; Thus, can excavate a plurality of associated entity of attribute to be associated more accurately, in like manner realize and excavate the corresponding user-specific attributes of entity (being attribute to be associated), thereby provide more comprehensively, detailed service meticulousr, more high-quality, for example, according to user-specific attributes to user's recommended entity; According to this device, can also excavate the incidence relation between any domain entities and user-specific attributes (being attribute to be associated), not be subject to the restriction of application, be widely used.
Should be appreciated that each several part of the present invention can realize with hardware, software, firmware or their combination.In the above-described embodiment, a plurality of steps or method can realize with being stored in storer and by software or the firmware of suitable instruction execution system execution.For example, if realized with hardware, the same in another embodiment, can realize by any one in following technology well known in the art or their combination: have for data-signal being realized to the discrete logic of the logic gates of logic function, the special IC with suitable combinational logic gate circuit, programmable gate array (PGA), field programmable gate array (FPGA) etc.
In the description of this instructions, the description of reference term " embodiment ", " some embodiment ", " example ", " concrete example " or " some examples " etc. means to be contained at least one embodiment of the present invention or example in conjunction with specific features, structure, material or the feature of this embodiment or example description.In this manual, the schematic statement of above-mentioned term is not necessarily referred to identical embodiment or example.And the specific features of description, structure, material or feature can be with suitable mode combinations in any one or more embodiment or example.
Although illustrated and described embodiments of the invention, those having ordinary skill in the art will appreciate that: in the situation that not departing from principle of the present invention and aim, can carry out multiple variation, modification, replacement and modification to these embodiment, scope of the present invention is limited by claim and equivalent thereof.
Claims (10)
1. a method of excavating attribute and entity associated relation, is characterized in that, comprises the following steps:
Obtain attribute to be associated;
According to described attribute to be associated, from a plurality of entities, obtain at least one and plant fructification; And
Obtain the associated entity of described at least one kind fructification, and described attribute to be associated is associated with the associated entity of described at least one kind fructification, described at least one kind fructification.
2. method according to claim 1, is characterized in that, described at least one kind fructification that obtains from a plurality of entities according to attribute to be associated specifically comprises:
From default entity storehouse, obtain described a plurality of entity;
From overall user sample cluster, obtain the associated user sample cluster with described attribute to be associated;
Obtain respectively a plurality of distributional difference values of described a plurality of entity in described associated user sample cluster; And
According to described a plurality of distributional difference values, described a plurality of entities are screened to obtain described at least one kind fructification.
3. method according to claim 2, is characterized in that, the described a plurality of distributional difference values of a plurality of entities in described associated user sample cluster of obtaining respectively specifically comprise:
Obtain respectively a plurality of first distribution proportions of a plurality of users relevant to described a plurality of entities in described overall user sample cluster;
Obtain respectively the second distribution proportion of a plurality of users relevant to described a plurality of entities in described associated user sample cluster; And
According to described the second distribution proportion and described the first distribution proportion, obtain described distributional difference value.
4. method according to claim 3, it is characterized in that, described overall user sample cluster is a plurality of, corresponding a plurality of network application services respectively, the distributional difference value that described in each, entity is corresponding is a plurality of, described according to a plurality of distributional difference values to described a plurality of entities screen to obtain described at least one plant fructification and also comprise:
According to default distributional difference value screening rule, described a plurality of entities are screened; Or,
Create distributional difference value sorter, and according to described distributional difference value sorter, described a plurality of entities are screened.
5. according to the method described in any one in claim 1 to 4, it is characterized in that, the associated entity of at least one kind fructification of described acquisition specifically comprises:
Obtain respectively described at least one plant fructification to the first incidence relation having between user's sample cluster of described attribute to be associated;
Obtain the associated entity group of user's sample cluster with described attribute to be associated, and user's sample cluster described in obtaining with described attribute to be associated is to the second incidence relation between described associated entity group;
According to described the first incidence relation and described the second incidence relation, obtain respectively described at least one kind fructification to described associated entity group's the 3rd incidence relation; And
According to described the 3rd incidence relation to associated entity described in each in described associated entity group screen to obtain described at least one plant the associated entity of fructification.
6. a device that excavates attribute and entity associated relation, is characterized in that, comprising:
Attribute acquisition module to be associated, for obtaining attribute to be associated;
Plant fructification acquisition module, for obtaining at least one according to described attribute to be associated from a plurality of entities, plant fructification;
Associated entity acquisition module, for obtaining the associated entity of described at least one kind fructification; And
Relating module, for being associated described attribute to be associated with the associated entity of described at least one kind fructification, described at least one kind fructification.
7. device according to claim 6, is characterized in that, described kind of fructification acquisition module comprises:
Entity acquiring unit, for obtaining described a plurality of entity from default entity storehouse;
Associated user sample cluster acquiring unit, for obtaining the associated user sample cluster with described attribute to be associated from overall user sample cluster;
Distributional difference value acquiring unit, for obtaining respectively described a plurality of entity in a plurality of distributional difference values of described associated user sample cluster; And
Screening unit, for screening to obtain described at least one kind fructification according to described a plurality of distributional difference values to described a plurality of entities.
8. device according to claim 7, it is characterized in that, described distributional difference value acquiring unit also specifically for: obtain respectively a plurality of first distribution proportions of a plurality of users relevant to described a plurality of entities in described overall user sample cluster, and obtain respectively the second distribution proportion of a plurality of users relevant to described a plurality of entities in described associated user sample cluster, and obtain described distributional difference value according to described the second distribution proportion and described the first distribution proportion.
9. device according to claim 8, it is characterized in that, described overall user sample cluster is a plurality of, corresponding a plurality of network application services respectively, the distributional difference value that described in each, entity is corresponding is a plurality of, describedly according to a plurality of distributional difference values, described a plurality of entities is also screened and is comprised:
According to default distributional difference value screening rule, described a plurality of entities are screened; Or,
Create distributional difference value sorter, and according to described distributional difference value sorter, described a plurality of entities are screened.
10. according to the device described in any one in claim 6 to 9, it is characterized in that, described associated entity acquisition module comprises:
The first incidence relation acquiring unit, for obtain respectively described at least one plant fructification to the first incidence relation having between user's sample cluster of described attribute to be associated;
The second incidence relation acquiring unit, for obtaining the associated entity group of user's sample cluster with described attribute to be associated, and user's sample cluster described in obtaining with described attribute to be associated is to the second incidence relation between described associated entity group;
The 3rd incidence relation acquiring unit, for obtaining respectively described at least one kind fructification to described associated entity group's the 3rd incidence relation according to described the first incidence relation and described the second incidence relation; And
Screening unit, for according to described the 3rd incidence relation to associated entity described in each of described associated entity group screen to obtain described at least one plant the associated entity of fructification.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310714291.6A CN103631970B (en) | 2013-12-20 | 2013-12-20 | The method and apparatus for excavating attribute and entity associated relation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310714291.6A CN103631970B (en) | 2013-12-20 | 2013-12-20 | The method and apparatus for excavating attribute and entity associated relation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103631970A true CN103631970A (en) | 2014-03-12 |
| CN103631970B CN103631970B (en) | 2017-08-18 |
Family
ID=50213011
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310714291.6A Active CN103631970B (en) | 2013-12-20 | 2013-12-20 | The method and apparatus for excavating attribute and entity associated relation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103631970B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105224642A (en) * | 2015-09-25 | 2016-01-06 | 百度在线网络技术（北京）有限公司 | The abstracting method of entity tag and device |
| CN105760491A (en) * | 2016-02-18 | 2016-07-13 | 中国科学院信息工程研究所 | Data modeling method and device based on equipment functions |
| CN107402933A (en) * | 2016-05-20 | 2017-11-28 | 富士通株式会社 | Entity polyphone disambiguation method and entity polyphone disambiguation equipment |
| CN107544992A (en) * | 2016-06-27 | 2018-01-05 | 阿里巴巴集团控股有限公司 | The method and apparatus of data analysis |
| CN108304493A (en) * | 2018-01-10 | 2018-07-20 | 深圳市腾讯计算机系统有限公司 | A kind of the hypernym method for digging and device of knowledge based collection of illustrative plates |
| CN108334632A (en) * | 2018-02-26 | 2018-07-27 | 深圳市腾讯计算机系统有限公司 | Entity recommends method, apparatus, computer equipment and computer readable storage medium |
| CN110188148A (en) * | 2019-05-23 | 2019-08-30 | 北京建筑大学 | Entity recognition method and device towards multimode heterogeneous characteristic |
| CN111047453A (en) * | 2019-12-04 | 2020-04-21 | 兰州交通大学 | Detection method and device for decomposing large-scale social network community based on high-order tensor |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101308493A (en) * | 2007-05-18 | 2008-11-19 | 亿览在线网络技术（北京）有限公司 | Entity relation exhibition method and system |
| CN102063433A (en) * | 2009-11-16 | 2011-05-18 | 华为技术有限公司 | Method and device for recommending related items |
| CN102915335A (en) * | 2012-09-17 | 2013-02-06 | 北京大学 | Information associating method based on user operation record and resource content |
| CN103425748A (en) * | 2013-07-19 | 2013-12-04 | 百度在线网络技术（北京）有限公司 | Method and device for mining document resource recommended words |
-
2013
- 2013-12-20 CN CN201310714291.6A patent/CN103631970B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101308493A (en) * | 2007-05-18 | 2008-11-19 | 亿览在线网络技术（北京）有限公司 | Entity relation exhibition method and system |
| CN102063433A (en) * | 2009-11-16 | 2011-05-18 | 华为技术有限公司 | Method and device for recommending related items |
| CN102915335A (en) * | 2012-09-17 | 2013-02-06 | 北京大学 | Information associating method based on user operation record and resource content |
| CN103425748A (en) * | 2013-07-19 | 2013-12-04 | 百度在线网络技术（北京）有限公司 | Method and device for mining document resource recommended words |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105224642A (en) * | 2015-09-25 | 2016-01-06 | 百度在线网络技术（北京）有限公司 | The abstracting method of entity tag and device |
| CN105224642B (en) * | 2015-09-25 | 2019-03-12 | 百度在线网络技术（北京）有限公司 | The abstracting method and device of entity tag |
| CN105760491A (en) * | 2016-02-18 | 2016-07-13 | 中国科学院信息工程研究所 | Data modeling method and device based on equipment functions |
| CN107402933A (en) * | 2016-05-20 | 2017-11-28 | 富士通株式会社 | Entity polyphone disambiguation method and entity polyphone disambiguation equipment |
| CN107544992A (en) * | 2016-06-27 | 2018-01-05 | 阿里巴巴集团控股有限公司 | The method and apparatus of data analysis |
| CN108304493A (en) * | 2018-01-10 | 2018-07-20 | 深圳市腾讯计算机系统有限公司 | A kind of the hypernym method for digging and device of knowledge based collection of illustrative plates |
| CN108304493B (en) * | 2018-01-10 | 2020-06-12 | 深圳市腾讯计算机系统有限公司 | Hypernym mining method and device based on knowledge graph |
| CN108334632A (en) * | 2018-02-26 | 2018-07-27 | 深圳市腾讯计算机系统有限公司 | Entity recommends method, apparatus, computer equipment and computer readable storage medium |
| CN108334632B (en) * | 2018-02-26 | 2021-03-23 | 深圳市腾讯计算机系统有限公司 | Entity recommendation method and device, computer equipment and computer-readable storage medium |
| CN110188148A (en) * | 2019-05-23 | 2019-08-30 | 北京建筑大学 | Entity recognition method and device towards multimode heterogeneous characteristic |
| CN111047453A (en) * | 2019-12-04 | 2020-04-21 | 兰州交通大学 | Detection method and device for decomposing large-scale social network community based on high-order tensor |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103631970B (en) | 2017-08-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103631970A (en) | Method and device for mining associated relationship between attributes and entities | |
| Fernandes et al. | Resolving galaxies in time and space-I. Applying STARLIGHT to CALIFA datacubes | |
| Guo et al. | A case study using visualization interaction logs and insight metrics to understand how analysts arrive at insights | |
| Surian et al. | Recommending people in developers' collaboration network | |
| CN106055617A (en) | Data pushing method and device | |
| CN102750336B (en) | Resource individuation recommendation method based on user relevance | |
| US10552390B2 (en) | Root cause analysis of performance problems | |
| CN107633019A (en) | A kind of page events acquisition method and device | |
| CN100354865C (en) | Fine-grained webpage information acquisition method | |
| CN102307315B (en) | User behavior analysis device in Internet protocol television (IPTV) system, and system for realizing analysis application | |
| CN103713989A (en) | Test case generating method and test case generating device for user terminal | |
| CN102724059A (en) | Website operation state monitoring and abnormal detection based on MapReduce | |
| CN102043716A (en) | Automatic software testing method based on business driving | |
| CN106326413A (en) | Personalized video recommending system and method | |
| Reinecke et al. | Phase-type fitting using HyperStar | |
| CN103164481A (en) | Recommendation method and system of video with largest rising trend | |
| CN108052608B (en) | Method and device for intelligently recommending university major according to high school course | |
| CN105913145A (en) | Data driving-based AB test method | |
| CN106204099A (en) | Based on Element-Level other advertising creative efficiency analysis method and device | |
| CN104933171A (en) | Method and device for associating data of interest point | |
| CN112446574B (en) | Product evaluation method, device, electronic equipment and storage medium | |
| CN103390067A (en) | Data processing method and device for internet entity analysis | |
| Trani et al. | WFCatalog: A catalogue for seismological waveform data | |
| Hawken et al. | Safer cities for women: Global and local innovations with open data and civic technology | |
| US11157267B1 (en) | Evaluation of dynamic relationships between application components |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |