CN104102712A - Method and device for entity identification - Google Patents
Method and device for entity identification Download PDFInfo
- Publication number
- CN104102712A CN104102712A CN201410337306.6A CN201410337306A CN104102712A CN 104102712 A CN104102712 A CN 104102712A CN 201410337306 A CN201410337306 A CN 201410337306A CN 104102712 A CN104102712 A CN 104102712A
- Authority
- CN
- China
- Prior art keywords
- result
- input message
- user
- attribute
- candidate
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
Abstract
The invention provides a method and a device for entity identification. The method includes: receiving input information, input by a user, of an to-be-identified entity, and acquiring a candidate result set according to the input information; when the candidate result set does not satisfy preset conditions, displaying output information to the user, and receiving input information re-input by the user according to the output information, and reacquiring a candidate result set according to the re-input input information till the candidate result set satisfies the preset conditions; taking results in the candidate result set satisfying the preset conditions as identification results to be displayed to the user. By the method, accuracy of entity identification can be improved, and user experience is improved.
Description
Technical field
The present invention relates to communication technical field, relate in particular to a kind of method and apparatus of identifying entity.
Background technology
When people run into unknown things, conventionally understand seeking way and understand this things.For example, unknown things is taken pictures, then adopt the image recognition series products search result relevant to this things, image recognition series products is finding after the result associated with this things conventionally, and preceding sequence several results are once returned to user; Or, user can Ai Te other people know result.
But aforesaid way is difficult to obtain result accurately.
Summary of the invention
The present invention is intended to solve at least to a certain extent one of technical matters in correlation technique.
For this reason, one object of the present invention is to propose a kind of method of identifying entity, and the method can improve the recognition accuracy to entity.
Another object of the present invention is to propose a kind of device of identifying entity.
For achieving the above object, the method for the identification entity that first aspect present invention embodiment proposes, comprising: receive the input message of the entity to be identified of user's input, and obtain candidate result collection according to described input message; Do not meet when pre-conditioned at described candidate result collection, show output information to described user, and, receive the input message that described user inputs again according to described output information, and again obtain candidate result collection according to the described input message of input again, until described candidate result collection meets pre-conditioned; Meet the concentrated result of pre-conditioned candidate result and show described user as recognition result described.
The method of the identification entity that first aspect present invention embodiment proposes, when pre-conditioned by not meeting at the candidate result collection obtaining according to input message, show output information to user, and receive the input message that user inputs again according to this output information, and obtain candidate result collection according to input message again, can realize many wheels mutual, and can incorporate more information in many wheels are mutual, because information that can be for reference is more, therefore can improve the recognition accuracy of entity, promote user and experience.
For achieving the above object, the device of identification entity that second aspect present invention embodiment proposes, comprising: acquisition module, for receiving the input message of entity to be identified of user's input, and obtains candidate result collection according to described input message; Processing module, when pre-conditioned for not meeting at described candidate result collection, show output information to described user, and, receive the input message that described user inputs again according to described output information, and again obtain candidate result collection according to the described input message of input again, until described candidate result collection meets pre-conditioned; Display module, for meeting the concentrated result of pre-conditioned candidate result and show described user as recognition result described.
The device of the identification entity that second aspect present invention embodiment proposes, when pre-conditioned by not meeting at the candidate result collection obtaining according to input message, show output information to user, and receive the input message that user inputs again according to this output information, and obtain candidate result collection according to input message again, can realize many wheels mutual, and can incorporate more information in many wheels are mutual, because information that can be for reference is more, therefore can improve the recognition accuracy of entity, promote user and experience.
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.
Brief description of the drawings
The present invention above-mentioned and/or additional aspect and advantage will become from the following description of the accompanying drawings of embodiments obviously and easily and understand, wherein:
Fig. 1 is the schematic flow sheet of a kind of data transmission method of embodiment of the present invention proposition;
Fig. 2 is that the one of inputting the interaction flow of flowers picture in the embodiment of the present invention is shown schematic diagram;
Fig. 3 is that the another kind of inputting the interaction flow of flowers picture in the embodiment of the present invention is shown schematic diagram;
Fig. 4 is that the one of inputting the interaction flow of pet picture in the embodiment of the present invention is shown schematic diagram;
Fig. 5 is that the another kind of inputting the interaction flow of pet picture in the embodiment of the present invention is shown schematic diagram;
Fig. 6 is the schematic flow sheet of the method for the identification entity of another embodiment of the present invention proposition;
Fig. 7 be in the embodiment of the present invention according to input message obtains first result not only for the moment the one of interaction flow show schematic diagram;
Fig. 8 be in the embodiment of the present invention according to input message obtains first result not only for the moment the another kind of interaction flow show schematic diagram;
Fig. 9 is a kind of schematic diagram of showing when input message is the entity of other types in the embodiment of the present invention;
Figure 10 is the another kind of schematic diagram of showing when input message is the entity of other types in the embodiment of the present invention;
Figure 11 shows schematic diagram only for the moment according to the result that input message obtains first in the embodiment of the present invention;
Figure 12 is the structural representation of the device of the identification entity of another embodiment of the present invention proposition;
Figure 13 is the structural representation of the device of the identification entity of another embodiment of the present invention proposition.
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.Be exemplary below by the embodiment being described with reference to the drawings, 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 all changes, amendment and the equivalent within the scope of spirit and the intension that falls into additional claims.
Fig. 1 is the schematic flow sheet of the method for the identification entity of one embodiment of the invention proposition, and the method comprises:
S11: receive the input message of the entity to be identified of user's input, and obtain candidate result collection according to described input message;
Wherein, the form of input message includes but not limited to: picture, and voice, or text, or, click behavior.Candidate answers option is for example clicked in click behavior.
Picture can comprise: the picture of selecting from local picture library, or, the current picture of taking pictures and obtaining.
For example, referring to Fig. 2, can input the input message 21 of picture form.
At service end building database in advance, obtaining after input message, the information in input message and database can be compared, find associated information as candidate result, the set of candidate result composition can be called candidate result collection.Concrete information comparison process can adopt normally used information searching technology, and the present embodiment does not limit this.
S12: do not meet when pre-conditioned at described candidate result collection, show output information to described user, and, receive the input message that described user inputs again according to described output information, and again obtain candidate result collection according to the described input message of input again, until described candidate result collection meets pre-conditioned.
Wherein, do not meet and pre-conditionedly can specifically refer to that candidate result concentrates and have result but not unique, now, can show output information to user, to obtain more information, to obtain unique result.
Referring to Fig. 2, in the time that candidate result is not unique, can show output information 22 as shown in Figure 2.
Showing that after output information, user can input again according to this output information.
Wherein, the input message of input comprises again: the information of user self input, for example, referring to Fig. 2, the input message 23 of input is that user inputs " May " again.Or,
The input message of input can also comprise again: the information that user selects in output information, for example, referring to Fig. 3, the input message 33 of input is that user selects " cochlear " in output information 32 again.Be understandable that, above-mentioned reciprocal process can exist repeatedly, for example, first carries out as shown in Figure 2 mutual, after user's input " May ", referring to Fig. 3, then carries out the reciprocal process of leaf-shaped.
S13: meet the concentrated result of pre-conditioned candidate result and show described user as recognition result described.
When carry out above-mentioned at least twice mutual after, if when the concentrated result of candidate result is unique, can exports and show user this unique result as recognition result.
For example, referring to Fig. 3, mutual through twice after, can obtain recognition result, recognition result can specifically comprise: entity title 34 and encyclopaedia introduce 35.
In above-mentioned reciprocal process, input message and/or output information can be: picture form, or, speech form, or textual form.For example, the input message 21 in Fig. 2 is picture forms, and input message 22 is textual forms again; Referring to Fig. 3, the title of recognition result is linguistic form, and the Baidu of recognition result introduces and comprises image content.
Optionally, receiving after user's input message, can first judge whether this input message belongs to default identification types, in the time belonging to default identification types, carrying out above-mentioned identification.
Default identification types can comprise: flowers, or, pet.
Above-mentioned Fig. 2 and Fig. 3 have provided the identification schematic diagram to flowers, similarly, adopt above-mentioned flow process also can realize the identification to pet.
For example, referring to Fig. 4, the first input message 41 of user's input is the picture of pet, afterwards in the time that the concentrated result of the candidate result obtaining according to the first input message is not unique, can show the first output information 42 as shown in Figure 4, user can input the second input message 43 again according to the first output information 42, in the time that the concentrated result of the candidate result obtaining according to the second input message 43 is not unique, can again show the second output information 44, afterwards, corresponding the second output information 44 can be selected the 3rd input message 45, through above-mentioned mutual after, can obtain recognition result as shown in Figure 5, recognition result comprises entity title 46 and Baidupedia 47.
In addition, in above-mentioned user's input process, in the time that output information comprises the problem of the attribute to selecting, problem " you saw it at several months " as shown in Figure 2, the answer of user's input can be the answer with clear and definite implication, for example " May ", or, the answer of user's input can be also the answer without clear and definite implication, for example, and " not knowing ", " unclear ", " having forgotten " this class answer.
When the present embodiment is pre-conditioned by not meeting at the candidate result collection obtaining according to input message, show output information to user, and receive the input message that user inputs again according to this output information, and obtain candidate result collection according to input message again, can realize many wheels mutual, and can incorporate more information in many wheels are mutual, because information that can be for reference is more, therefore can improve the recognition accuracy of entity, promote user and experience.
Fig. 6 is the schematic flow sheet of the method for the identification entity of another embodiment of the present invention proposition, comprising:
S601: resolve the input message receiving.
Wherein, input message can be picture form, speech form or textual form.
S602: judge whether this input message belongs to default identification types, if so, carries out S603, otherwise, carry out S610.
Wherein, default identification types can comprise: flowers or pet, and in the time that input message is flowers information or pet information, determines that input message belongs to default identification types, otherwise do not belong to.
S603: obtain candidate result collection.
Wherein, can obtain candidate result collection from service end, service end can be from the database of setting up in advance, according to the preset algorithm retrieval information associated with input message, the associated information retrieving is defined as to candidate result, and optional, candidate result can form candidate result collection.
S604: judge candidate result concentrates whether there is result, if so, carries out S605, otherwise, carry out S614.
S605: judge that whether the concentrated result of candidate result is unique, if so, carry out S613, otherwise, carry out S606.
S606: judge whether the concentrated result of candidate result does not inquire attribute, if so, carries out S607, otherwise, carry out S612.
Wherein, can set in advance the attribute to each result, for example, for flowers class, can set a property and comprise: leaf-shaped, growth place, growth time etc.; Or, for pets, can set a property and comprise: ear shape, pet color etc.
In addition, can also determine according to default attribute prioritization algorithm the priority of each attribute, in addition, attribute priority can dynamic change, while identifying different flowers, can adopt different attribute prioritization algorithms, while making to identify different flowers, the priority difference of the problem of asking, the candidate answers of each problem is also different.Corresponding a certain class is determined after flowers, can be according to attribute priority query attribute corresponding to such flowers, for example, for certain class flowers, if the priority of the priority ratio leaf-shaped of current definite growth time is high, so, in the time of inquiry, can first inquire growth time, in the time unique result can not being obtained according to growth time, then inquire leaf-shaped.
S607: obtain the property value of not inquiring attribute.
The property value that can preserve in advance every kind of attribute in database, for example, the property value of growth time can comprise: January, October, November.
S608: to not inquiring that attribute sorts, by attribute and corresponding property value composition candidate value list.
Wherein, can sort to attribute according to the priority of attribute.And property value also can arrange priority, corresponding each attribute, property value also can be according to prioritization.
S609: show that to user attribute that priority is the highest is as problem, and the property value of the default number of this attribute is showed to user as candidate item.
Wherein, the attribute that priority is the highest refers to does not inquire the highest attribute of attribute medium priority.
The property value of default number can be to select according to the priority of property value.
Suppose not inquire that in attribute, the attribute of limit priority is petal color, the default number of property value is three, property value is respectively green, purple and white according to first three color of priority queueing,, referring to Fig. 7, can show the output information that comprises problem 71 and candidate answers 72 as shown in Figure 7.
Showing after output information to user, user can input answer that should problem as the input message of input again, can repeat afterwards S601 and subsequent step thereof.
Referring to Fig. 8, according to above-mentioned interaction flow, can obtain recognition result, recognition result comprises that entity title 81 and encyclopaedia introduce 82.
S610: judge whether to find other entities according to this input message, if so, carry out S611, otherwise, carry out S614.
Wherein, other entities refer to the entity that is not to belong to this default identification types.
Concrete, can be in database can also other types entity, for example, the entity of non-flowers class and the entity of non-pets, the entity that similarity can be greater than to predetermined threshold value is defined as final recognition result.
S611: the identification of other types is returned as recognition result.Afterwards, carry out S615.
In addition, can also return to information, described information is for pointing out user to input the information of entity corresponding to described default identification types.
For example,, referring to Fig. 9, in the time that input message 91 is the entity of other types, the output information 92 of returning comprises recognition result and information, similarly, and referring to Figure 10, in the time that input message 101 is the entity of other types, the output information 102 of returning comprises recognition result and information.
S612: judge whether history of existence result, if so, carry out S613, otherwise, carry out S614.
Wherein, historical results refer to according on take turns the concentrated result of candidate result obtaining alternately, this result is not unique.
S613: return to recognition result, this recognition result comprises entity title and encyclopaedia introduction.
Concrete, when the concentrated result of the candidate result of obtaining is unique, using this unique result as recognition result.
For example, referring to Figure 11, when the concentrated result of the candidate result obtaining according to input message 111 is while being unique, can export recognition result as shown in figure 11, this recognition result comprises that entity title 112 and encyclopaedia introduce 113.
Or, in the time having historical results, using on take turns at least two results that the candidate result that obtains when mutual concentrates as recognition result, the output of each result is shown also can comprise entity title and encyclopaedia introduction.
S614: return without results suggest.
S615: finish.
Optionally, the method for the present embodiment can realize in independent product, can also be applied in other products, and for example, the method can be in independent application program, also can be applied in the platforms such as mhkc, Baidu are known, micro-letter, microblogging.
The present embodiment, by many wheel interactive identification entities, can improve the overall accuracy rate of identification, and in whole identifying, provides more humane user to experience; The present embodiment also can be identified and be pointed out in the time of the entity of input other types, can in the time that user deliberately disorderly claps, expose user, and guides its recurrence scene with friendly official documents and correspondence; The present embodiment is by being combined with other products, the mutual scope that can extend one's service.
Figure 12 is the structural representation of the device of the identification entity of another embodiment of the present invention proposition, and this device 120 comprises acquisition module 121, processing module 122 and display module 123.
Acquisition module 121 is for receiving the input message of entity to be identified of user input, and obtains candidate result collection according to described input message.
Wherein, the form of input message includes but not limited to: picture, and voice, or text, or, click behavior, for example the click behavior to candidate answers.
Picture can comprise: the picture of selecting from local picture library, or, the current picture of taking pictures and obtaining.
For example, referring to Fig. 2, can input the input message 21 of picture form.
At service end building database in advance, obtaining after input message, the information in input message and database can be compared, find associated information as candidate result, the set of candidate result composition can be called candidate result collection.Concrete information comparison process can adopt normally used information searching technology, and the present embodiment does not limit this.
When processing module 122 is pre-conditioned for not meeting at described candidate result collection, show output information to described user, and, receive the input message that described user inputs again according to described output information, and again obtain candidate result collection according to the described input message of input again, until described candidate result collection meets pre-conditioned.
Wherein, do not meet and pre-conditionedly can specifically refer to that candidate result concentrates and have result but not unique, now, can show output information to user, to obtain more information, to obtain unique result.
Referring to Fig. 2, in the time that candidate result is not unique, can show output information 22 as shown in Figure 2.
Showing that after output information, user can input again according to this output information.
Wherein, the input message of input comprises again: the information of user self input, for example, referring to Fig. 2, the input message 23 of input is that user inputs " May " again.Or,
The input message of input can also comprise again: the information that user selects in output information, for example, referring to Fig. 3, the input message 33 of input is that user selects " cochlear " in output information 32 again.
Be understandable that, above-mentioned reciprocal process can exist repeatedly, for example, first carries out as shown in Figure 2 mutual, after user's input " May ", referring to Fig. 3, then carries out the reciprocal process of leaf-shaped.
Display module 123 is for meeting the concentrated result of pre-conditioned candidate result and show described user as recognition result described.
When carry out above-mentioned at least twice mutual after, if when the concentrated result of candidate result is unique, can exports and show user this unique result as recognition result.
For example, referring to Fig. 3, mutual through twice after, can obtain recognition result, recognition result can specifically comprise: entity title 34 and encyclopaedia introduce 35.
In above-mentioned reciprocal process, input message and/or output information can be: picture form, or, speech form, or textual form, or click behavior.For example, the input message 21 in Fig. 2 is picture forms, and input message 22 is textual forms again; Referring to Fig. 3, the title of recognition result is linguistic form, and the Baidu of recognition result introduces and comprises image content.
Optionally, receiving after user's input message, can first judge whether this input message belongs to default identification types, in the time belonging to default identification types, carrying out above-mentioned identification.
Default identification types can comprise: flowers, or, pet.
Above-mentioned Fig. 2 and Fig. 3 have provided the identification schematic diagram to flowers, similarly, adopt above-mentioned flow process also can realize the identification to pet.
For example, referring to Fig. 4, the first input message 41 of user's input is the picture of pet, afterwards in the time that the concentrated result of the candidate result obtaining according to the first input message is not unique, can show the first output information 42 as shown in Figure 4, user can input the second input message 43 again according to the first output information 42, in the time that the concentrated result of the candidate result obtaining according to the second input message 43 is not unique, can again show the second output information 44, afterwards, corresponding the second output information 44 can be selected the 3rd input message 45, through above-mentioned mutual after, can obtain recognition result as shown in Figure 5, recognition result comprises entity title 46 and Baidupedia 47.
In addition, in above-mentioned user's input process, in the time that output information comprises the problem of the attribute to selecting, problem " you saw it at several months " as shown in Figure 2, the answer of user's input can be the answer with clear and definite implication, for example " May ", or, the answer of user's input can be also the answer without clear and definite implication, for example, and " not knowing ", " unclear ", " having forgotten " this class answer.
In an embodiment, as shown in figure 13, described device 120 also comprises the first judge module 124, for judging whether described input message belongs to default identification types, to in the time that described input message belongs to default identification types, obtain candidate result collection according to described input message.
Wherein, default identification types can comprise: flowers or pet, and in the time that input message is flowers information or pet information, determines that input message belongs to default identification types, otherwise do not belong to.
In an embodiment, acquisition module 31 can obtain candidate result collection from service end, service end can be from the database of setting up in advance, according to the preset algorithm retrieval information associated with input message, the associated information retrieving is defined as to candidate result, optionally, candidate result can form candidate result collection.
In an embodiment, described processing module 122 is specifically for judging whether the concentrated result of described candidate result does not inquire attribute; In the time not inquiring attribute, select an attribute described in not inquiring attribute; Show output information to described user, described output information comprises the problem of the attribute to described selection.
Wherein, can set in advance the attribute to each result, for example, for flowers class, can set a property and comprise: leaf-shaped, growth place, growth time etc.; Or, for pets, can set a property and comprise: ear shape, pet color etc.
In addition, the priority of each attribute can also be set in the time arranging, according to priority query attribute, for example, for flowers class, the priority of priority ratio leaf-shaped that growth time can be set is high, so, in the time of inquiry, can first inquire growth time, in the time can not obtaining unique result according to growth time, then inquire leaf-shaped.
In another embodiment, also comprise candidate answers in described output information, described processing module 122 is also specifically for obtaining at least one property value of attribute of described selection; Described at least one property value is defined as to described candidate answers.
The property value that can preserve in advance every kind of attribute in database, for example, the property value of growth time can comprise: January, October, November.
Wherein, can sort to attribute according to the priority of attribute.And property value also can arrange priority, corresponding each attribute, property value also can be according to prioritization.
Wherein, the attribute that priority is the highest refers to does not inquire the highest attribute of attribute medium priority.
The property value of default number can be to select according to the priority of property value.
Suppose not inquire that in attribute, the attribute of limit priority is petal color, the default number of property value is three, property value is respectively green, purple and white according to first three color of priority queueing,, referring to Fig. 7, can show the output information that comprises problem 71 and candidate answers 72 as shown in Figure 7.
Referring to Fig. 8, according to above-mentioned interaction flow, can obtain recognition result, recognition result comprises that entity title 81 and encyclopaedia introduce 82.
In another embodiment, the answer that described processing module 122 is also selected in described candidate answers specifically for receiving described user; Or, receive the answer to described problem of described user's input.
Showing after output information to user, user can input answer that should problem as the input message of input again, can repeat afterwards above-mentioned flow process and obtain showing result.
In another embodiment, when display module 123 is not inquired attribute specifically for judging in described processing module 122, using on the historical results of taking turns when mutual show described user as recognition result.In the time having historical results, using on take turns at least two results that the candidate result that obtains when mutual concentrates as recognition result, the output of each result is shown also can comprise entity title and encyclopaedia introduction.
In another embodiment, as shown in figure 13, described device 120 also comprises:
The second judge module 125, when do not belong to default identification types in described input message, judges whether to exist the entity of the other types corresponding with described input message;
Wherein, other entities refer to the entity that is not to belong to this default identification types.
Concrete, can be in database can also other types entity, for example, the entity of non-flowers class and the entity of non-pets, the entity that similarity can be greater than to predetermined threshold value is defined as final recognition result.
Reminding module 126, if there is the entity of the other types corresponding with described input message for the second judge module judgement, return to described user using the entity of described other types as recognition result, and show information to described user, described information is for pointing out user to input the information of entity corresponding to described default identification types.
For example,, referring to Fig. 9, in the time that input message 91 is the entity of other types, the output information 92 of returning comprises recognition result and information, similarly, and referring to Figure 10, in the time that input message 101 is the entity of other types, the output information 102 of returning comprises recognition result and information.
In another embodiment, as shown in figure 13, described device 120 also comprises:
The 3rd judge module 127, for judging described candidate result concentrates whether there is result;
The 4th judge module 128, concentrates while having result for judge described candidate result at described the 3rd judge module, judges that whether described result is unique;
Determination module 129, for judging described result at described four module when not unique, determines that described candidate result collection does not meet pre-conditioned.
Described display module 123 specifically for:
If judge describedly when unique at described the 4th judge module 128, described unique result is showed to described user as recognition result.
For example, referring to Figure 11, when the concentrated result of the candidate result obtaining according to input message 111 is while being unique, can export recognition result as shown in figure 11, this recognition result comprises that entity title 112 and encyclopaedia introduce 113.
Optionally, the method for the present embodiment can realize in independent product, can also be applied in other products, and for example, the method can be in independent application program, also can be applied in the platforms such as mhkc, Baidu are known, micro-letter, microblogging.
When the present embodiment is pre-conditioned by not meeting at the candidate result collection obtaining according to input message, show output information to user, and receive the input message that user inputs again according to this output information, and obtain candidate result collection according to input message again, can realize many wheels mutual, and can incorporate more information in many wheels are mutual, because information that can be for reference is more, therefore can improve the recognition accuracy of entity, promote user and experience.
It should be noted that, in description of the invention, term " first ", " second " etc. are only for describing object, and can not be interpreted as instruction or hint relative importance.In addition,, in description of the invention, except as otherwise noted, the implication of " multiple " is two or more.
Any process of otherwise describing in process flow diagram or at this or method are described 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 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.
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, multiple steps or method can realize with being stored in software or the firmware carried out in storer and by suitable instruction execution system.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: there is the discrete logic for data-signal being realized to the logic gates of logic function, there is the special IC of suitable combinational logic gate circuit, programmable gate array (PGA), field programmable gate array (FPGA) etc.
Those skilled in the art are appreciated that realizing all or part of step that above-described embodiment method carries is can carry out the hardware that instruction is relevant by program to complete, described program can be stored in a kind of computer-readable recording medium, this program, in the time carrying out, comprises step of embodiment of the method one or a combination set of.
In addition, the each functional unit in each embodiment of the present invention can be integrated in a processing module, can be also that the independent physics of unit exists, and also can be integrated in a module two or more unit.Above-mentioned integrated module both can adopt the form of hardware to realize, and also can adopt the form of software function module to realize.If described integrated module realizes and during as production marketing independently or use, also can be stored in a computer read/write memory medium using the form of software function module.
The above-mentioned storage medium of mentioning can be ROM (read-only memory), disk or CD 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 specific features, structure, material or the feature of description can be with suitable mode combination in any one or more embodiment or example.
Although illustrated and described embodiments of the invention above, be understandable that, above-described embodiment is exemplary, can not be interpreted as limitation of the present invention, and those of ordinary skill in the art can change above-described embodiment within the scope of the invention, amendment, replacement and modification.
Claims (20)
1. a method of identifying entity, is characterized in that, comprising:
Receive the input message of the entity to be identified of user's input, and obtain candidate result collection according to described input message;
Do not meet when pre-conditioned at described candidate result collection, show output information to described user, and, receive the input message that described user inputs again according to described output information, and again obtain candidate result collection according to the described input message of input again, until described candidate result collection meets pre-conditioned;
Meet the concentrated result of pre-conditioned candidate result and show described user as recognition result described.
2. method according to claim 1, is characterized in that, the form of described input message and/or described output information comprises:
Picture, or, voice, or, text, or, click behavior.
3. method according to claim 1, is characterized in that, described obtain candidate result collection according to described input message before, described method also comprises:
Judge whether described input message belongs to default identification types, in the time that described input message belongs to default identification types, obtain candidate result collection according to described input message.
4. method according to claim 3, is characterized in that, described judge whether described input message belongs to default identification types after, described method also comprises:
In the time that described input message does not belong to default identification types, judge whether to exist the entity of the other types corresponding with described input message;
If existed, return to described user using the entity of described other types as recognition result, and show information to described user, described information is for pointing out user to input the information of entity corresponding to described default identification types.
5. method according to claim 1, is characterized in that, described described candidate result collection do not meet pre-conditioned before, described method also comprises:
Judge described candidate result concentrates whether there is result;
If there is result, judge that whether described result is unique;
If not unique, determine that described candidate result collection does not meet pre-conditioned.
6. method according to claim 5, is characterized in that, described judge that whether described result unique after, described method also comprises:
If unique, described unique result is showed to described user as recognition result.
7. method according to claim 1, is characterized in that, describedly shows output information to described user, comprising:
Judge whether the concentrated result of described candidate result does not inquire attribute;
In the time not inquiring attribute, select an attribute described in not inquiring attribute;
Show output information to described user, described output information comprises the problem of the attribute to described selection.
8. method according to claim 7, is characterized in that, also comprises candidate answers in described output information, described described select an attribute in not inquiring attribute after, described method also comprises:
Obtain at least one property value of the attribute of described selection;
Described at least one property value is defined as to described candidate answers.
9. method according to claim 8, is characterized in that, the input message that the described user of described reception inputs again according to described output information, comprising:
Receive the answer that described user selects in described candidate answers; Or,
Receive the answer to described problem of described user's input, wherein, described answer comprises the answer that has the answer of clear and definite implication or do not have clear and definite implication.
10. method according to claim 7, is characterized in that, described judge whether the concentrated result of described candidate result does not inquire attribute after, described method also comprises:
In the time not inquiring attribute, using on the historical results of taking turns when mutual show described user as recognition result.
Identify the device of entity for 11. 1 kinds, it is characterized in that, comprising:
Acquisition module, for receiving the input message of entity to be identified of user input, and obtains candidate result collection according to described input message;
Processing module, when pre-conditioned for not meeting at described candidate result collection, show output information to described user, and, receive the input message that described user inputs again according to described output information, and again obtain candidate result collection according to the described input message of input again, until described candidate result collection meets pre-conditioned;
Display module, for meeting the concentrated result of pre-conditioned candidate result and show described user as recognition result described.
12. devices according to claim 11, is characterized in that, the form of described input message and/or described output information comprises:
Picture, or, voice, or, text, or, click behavior.
13. devices according to claim 11, is characterized in that, described device also comprises:
The first judge module, for judging whether described input message belongs to default identification types, in the time that described input message belongs to default identification types, obtain candidate result collection according to described input message.
14. devices according to claim 13, is characterized in that, described device also comprises:
The second judge module, when do not belong to default identification types in described input message, judges whether to exist the entity of the other types corresponding with described input message;
Reminding module, if there is the entity of the other types corresponding with described input message for the second judge module judgement, return to described user using the entity of described other types as recognition result, and show information to described user, described information is for pointing out user to input the information of entity corresponding to described default identification types.
15. devices according to claim 11, is characterized in that, described device also comprises:
The 3rd judge module, for judging described candidate result concentrates whether there is result;
The 4th judge module, concentrates while having result for judge described candidate result at described the 3rd judge module, judges that whether described result is unique;
Determination module, for judging described result at described four module when not unique, determines that described candidate result collection does not meet pre-conditioned.
16. devices according to claim 15, is characterized in that, described display module specifically for:
If judge describedly when unique at described the 4th judge module, described unique result is showed to described user as recognition result.
17. devices according to claim 11, is characterized in that, described processing module specifically for:
Judge whether the concentrated result of described candidate result does not inquire attribute;
In the time not inquiring attribute, select an attribute described in not inquiring attribute;
Show output information to described user, described output information comprises the problem of the attribute to described selection.
18. devices according to claim 17, is characterized in that, also comprise candidate answers in described output information, described processing module also specifically for:
Obtain at least one property value of the attribute of described selection;
Described at least one property value is defined as to described candidate answers.
19. devices according to claim 18, is characterized in that, described processing module also specifically for:
Receive the answer that described user selects in described candidate answers; Or,
Receive the answer to described problem of described user's input, wherein, described answer comprises the answer that has the answer of clear and definite implication or do not have clear and definite implication.
20. devices according to claim 17, is characterized in that, described display module specifically for:
Judge while not inquiring attribute in described processing module, using on the historical results of taking turns when mutual show described user as recognition result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410337306.6A CN104102712A (en) | 2014-07-15 | 2014-07-15 | Method and device for entity identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410337306.6A CN104102712A (en) | 2014-07-15 | 2014-07-15 | Method and device for entity identification |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104102712A true CN104102712A (en) | 2014-10-15 |
Family
ID=51670866
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410337306.6A Pending CN104102712A (en) | 2014-07-15 | 2014-07-15 | Method and device for entity identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104102712A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105934760A (en) * | 2014-01-24 | 2016-09-07 | 微软技术许可有限责任公司 | Adaptable image search with computer vision assistance |
| CN110334200A (en) * | 2019-07-11 | 2019-10-15 | 深圳市仝智科技有限公司 | A kind of generation method, device and the storage medium in problem answers library |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8548915B2 (en) * | 2006-12-21 | 2013-10-01 | Support Machines Ltd. | Method and computer program product for providing a response to a statement of a user |
| CN103366003A (en) * | 2013-07-19 | 2013-10-23 | 百度在线网络技术（北京）有限公司 | Method and device based on user feedback optimizing search result |
| CN103760991A (en) * | 2014-01-13 | 2014-04-30 | 北京搜狗科技发展有限公司 | Physical input method and physical input device |
| CN103914513A (en) * | 2014-01-13 | 2014-07-09 | 北京搜狗科技发展有限公司 | Entity input method and device |
| CN103914548A (en) * | 2014-04-10 | 2014-07-09 | 北京百度网讯科技有限公司 | Information searching method and information searching device |
-
2014
- 2014-07-15 CN CN201410337306.6A patent/CN104102712A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8548915B2 (en) * | 2006-12-21 | 2013-10-01 | Support Machines Ltd. | Method and computer program product for providing a response to a statement of a user |
| CN103366003A (en) * | 2013-07-19 | 2013-10-23 | 百度在线网络技术（北京）有限公司 | Method and device based on user feedback optimizing search result |
| CN103760991A (en) * | 2014-01-13 | 2014-04-30 | 北京搜狗科技发展有限公司 | Physical input method and physical input device |
| CN103914513A (en) * | 2014-01-13 | 2014-07-09 | 北京搜狗科技发展有限公司 | Entity input method and device |
| CN103914548A (en) * | 2014-04-10 | 2014-07-09 | 北京百度网讯科技有限公司 | Information searching method and information searching device |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105934760A (en) * | 2014-01-24 | 2016-09-07 | 微软技术许可有限责任公司 | Adaptable image search with computer vision assistance |
| CN105934760B (en) * | 2014-01-24 | 2019-09-03 | 微软技术许可有限责任公司 | It is searched for using the adaptive topography of computer vision auxiliary |
| CN110334200A (en) * | 2019-07-11 | 2019-10-15 | 深圳市仝智科技有限公司 | A kind of generation method, device and the storage medium in problem answers library |
| CN110334200B (en) * | 2019-07-11 | 2021-07-27 | 深圳市仝智科技有限公司 | Method and device for generating question answer library and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103853842B (en) | A kind of automatic question-answering method and system | |
| CN101128824B (en) | Location extraction | |
| US20160379106A1 (en) | Human-computer intelligence chatting method and device based on artificial intelligence | |
| CN110147722A (en) | A kind of method for processing video frequency, video process apparatus and terminal device | |
| CN102609434A (en) | Face recognition using social data | |
| CN105959374A (en) | Data recommendation method and data recommendation equipment | |
| CN104216965A (en) | Information recommendation method and device | |
| CN107590267A (en) | Information-pushing method and device, terminal and readable storage medium storing program for executing based on picture | |
| CN111680133A (en) | Live broadcast question and answer method and device | |
| CN111400607A (en) | Search content output method and device, computer equipment and readable storage medium | |
| EP3822814A2 (en) | Human-machine interaction method and apparatus based on neural network | |
| CN106789559A (en) | Information processing method, device and system for wechat public platform | |
| CN111931859B (en) | Multi-label image recognition method and device | |
| CN110825968A (en) | Information pushing method and device, storage medium and computer equipment | |
| CN110046231B (en) | Customer service information processing method, server and system | |
| CN106227510A (en) | Method and device is recommended in application | |
| CN104462590A (en) | Information searching method and device | |
| US11620453B2 (en) | System and method for artificial intelligence driven document analysis, including searching, indexing, comparing or associating datasets based on learned representations | |
| WO2018128718A1 (en) | System and method of generating a semantic representation of a target image for an image processing operation | |
| CN105677636A (en) | Information processing method and device for intelligent question-answering system | |
| US20160267425A1 (en) | Data processing techniques | |
| CN110059172B (en) | Method and device for recommending answers based on natural language understanding | |
| CN104268504B (en) | Image identification method and device | |
| CN104102712A (en) | Method and device for entity identification | |
| CN111737608B (en) | Method and device for ordering enterprise information retrieval results |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20141015 |