CN105095195A - Method and system for human-machine questioning and answering based on knowledge graph - Google Patents
Method and system for human-machine questioning and answering based on knowledge graph Download PDFInfo
- Publication number
- CN105095195A CN105095195A CN201510383452.7A CN201510383452A CN105095195A CN 105095195 A CN105095195 A CN 105095195A CN 201510383452 A CN201510383452 A CN 201510383452A CN 105095195 A CN105095195 A CN 105095195A
- Authority
- CN
- China
- Prior art keywords
- knowledge
- point
- knowledge point
- entity
- mapping
- 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
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a method and a system for human-machine questioning and answering based on a knowledge graph. In the method for the human-machine questioning and answering based on the knowledge graph provided by the invention, word segmentation is carried out to a sentence input by a user when the sentence input by the user is input; entity extraction is carried out to words obtained by the word segmentation; knowledge inference is carried out to entity information obtained by the entity extraction through application of the knowledge graph; and according to results of the knowledge inference, a feedback is made to the sentence input by the user, so that an answer fed back by the human-machine questioning and answering will become more accurate, a question proposed by the user can be answered in a targeted manner, and the user's satisfaction degree can be increased.
Description
Technical field
The present invention relates to the data processing being applicable to nan-machine interrogation, particularly nan-machine interrogation's method and system of knowledge based collection of illustrative plates.
Background technology
Along with the development of internet, applications, nan-machine interrogation's system is introduced in its website or APP by a lot of enterprise, public institution or functional government departments, to assist or to replace the consulting by artificial multiplexing family back and forth.
Mostly existing nan-machine interrogation's system is that keyword in the problem by extracting user is as knowledge point, and carries out man-to-man entity at data store internal and map and find out the respective items of knowledge point, then using respective items as answer feedback to user.Existing nan-machine interrogation's system only establishes man-to-man relationship map net to knowledge point, contacting between knowledge point and knowledge point is very weak, cannot carry out knowledge-based reasoning, and therefore its answer feeding back to user is often inaccurate, even lack specific aim, do not give a direct answer to a question.
Summary of the invention
In view of above-mentioned, the knowledge reasoning of knowledge based collection of illustrative plates is introduced nan-machine interrogation by the present invention, can carry out combing and the foundation of knowledge better, and then makes the representation of knowledge of machine perception complexity, thus accurately can answer user the problem proposed targetedly.
According to an aspect of the present invention, provide a kind of nan-machine interrogation's method of knowledge based collection of illustrative plates, comprising: receive the statement of user's input and participle is carried out to described statement; Entity extraction is carried out to the word obtained by described participle; Knowledge mapping is utilized to carry out knowledge reasoning to extracting by described entity the entity information obtained; And according to the result of described knowledge reasoning, feedback is provided to the statement that described user inputs.
Nan-machine interrogation's method according to embodiments of the invention, preferably, in the step of described participle, by segmentation methods, the statement that user inputs is cut, retain the stop words of relation belonging to Feature Words, and remove remaining stop words and redundant information.
Nan-machine interrogation's method according to embodiments of the invention, preferably, in the step that described entity extracts, the entity attribute of the word obtained is utilized to mark it, wherein, described entity attribute comprises the part of speech of institute's predicate, the dictionary lexical or textual analysis of institute's predicate or the product classification corresponding to institute's predicate.
Nan-machine interrogation's method according to embodiments of the invention, preferably, described knowledge mapping comprises the relation between knowledge point and knowledge point.
Nan-machine interrogation's method according to embodiments of the invention, preferably, when comprising an entity information, in the step of described knowledge reasoning, searches the knowledge point corresponding to described entity information in described knowledge mapping; After the described knowledge point of acquisition, terminate the step of described reasoning, and provide the positional information in described knowledge mapping with the content information corresponding to described knowledge point and described knowledge point; And when not finding described knowledge point when traveling through described knowledge mapping, terminating the step of described reasoning, and providing the feedback without accordingly result.
Nan-machine interrogation's method according to embodiments of the invention, preferably, when comprising many entity informations, in the step of described knowledge reasoning, a) searches the first knowledge point corresponding in described entity information in described knowledge mapping; B) by the out-degree point of described first knowledge point, utilize described first knowledge point and correspond to the relation between another the second knowledge point in described entity information, in described knowledge mapping, search described second knowledge point; C) repeat above-mentioned b step, search until completed for the whole knowledge point corresponded in described entity information, terminate the step of described reasoning, and provide and the content information corresponding to described knowledge point; And d) repeat above-mentioned b step, when not finding the knowledge point that will search when traveling through described knowledge mapping, terminating the step of described reasoning, and providing the feedback without accordingly result.
According to another aspect of the present invention, additionally provide a kind of nan-machine interrogation's system performing knowledge based collection of illustrative plates, described system comprises: word-dividing mode, carries out participle for the statement inputted user; Entity abstraction module, for carrying out entity extraction to the word obtained by described participle; Knowledge mapping module, for the relation between stored knowledge point and knowledge point; Knowledge reasoning module, carries out knowledge reasoning for utilizing described knowledge mapping to extracting by described entity the entity information obtained; And output module, for the result according to described knowledge reasoning, feedback is provided to the statement that described user inputs.
Nan-machine interrogation's system according to embodiments of the invention, preferably, described word-dividing mode is cut the statement that user inputs by segmentation methods, retains the stop words of relation belonging to Feature Words, and removes remaining stop words and redundant information.
Nan-machine interrogation's system according to embodiments of the invention, preferably, described entity abstraction module utilizes the entity attribute of the word obtained to mark it, and wherein, described entity attribute comprises the part of speech of institute's predicate, the dictionary lexical or textual analysis of institute's predicate or the product classification corresponding to institute's predicate.
Nan-machine interrogation's system according to embodiments of the invention, preferably, described knowledge mapping module is by arranging independently in-degree point and out-degree point for each described knowledge point, the knowledge relation of the classification between described knowledge point is set up according to identical in-degree point and out-degree point, between described knowledge point, set up knowledge non-directed graph, carry out the relation between stored knowledge point and knowledge point.
Nan-machine interrogation's system according to embodiments of the invention, preferably, when comprising an entity information, described knowledge reasoning module, when obtaining the knowledge point corresponding to described entity information by searching in described knowledge mapping, provides the positional information in described knowledge mapping with the content information corresponding to described knowledge point and described knowledge point; And when not finding described knowledge point when traveling through described knowledge mapping, provide the feedback without accordingly result.
Nan-machine interrogation's system according to embodiments of the invention, preferably, when comprising many entity informations, described knowledge reasoning module searches the first knowledge point corresponding in described entity information in described knowledge mapping; By the out-degree point of described first knowledge point, utilize described first knowledge point and correspond to the relation between another the second knowledge point in described entity information, in described knowledge mapping, search described second knowledge point; When repetition above-mentioned steps, until for the whole knowledge point corresponded in described entity information completed search time, provide and the content information corresponding to described knowledge point; And when not finding when traveling through described knowledge mapping the knowledge point that will search, provide the feedback without accordingly result.
According to another aspect of the present invention, provide a kind of nan-machine interrogation's system of knowledge based collection of illustrative plates, described system comprises: load module, for receiving the statement of user's input; Word-dividing mode, carries out participle for the statement inputted user; Entity abstraction module, for carrying out entity extraction to the word obtained by described participle; Knowledge mapping module, for the relation between stored knowledge point and knowledge point; Knowledge reasoning module, carries out knowledge reasoning for utilizing described knowledge mapping to extracting by described entity the entity information obtained; And output module, for the result according to described knowledge reasoning, feedback is provided to the statement that described user inputs.
The present invention by introducing knowledge reasoning and in conjunction with data processing, achieve the accurate analysis to the problem that user proposes, and can answer user's problem proposed targetedly in nan-machine interrogation, thus reaches the effect promoting user satisfaction.
Accompanying drawing explanation
Accompanying drawing illustrates embodiments of the invention, and is used from instructions one and explains principle of the present invention.In the accompanying drawings:
Fig. 1 is the exemplary plot of the overall process process of nan-machine interrogation's method according to knowledge based collection of illustrative plates of the present invention;
Fig. 2 is the example flow diagram of the nan-machine interrogation's method according to knowledge based collection of illustrative plates of the present invention;
Fig. 3 is the graph of a relation of the example according to knowledge mapping of the present invention;
Fig. 4 is the example block diagram of the nan-machine interrogation's system according to knowledge based collection of illustrative plates of the present invention.
Embodiment
Below in conjunction with accompanying drawing, the preferred implementation of the application is described in detail.
In order to the facility illustrated, at this first using " it is different what i Phone and Samsung mobile phone have " as the example of user's read statement.In existing nan-machine interrogation's system, conventional method is: first extract the keyword in above-mentioned statement, " i Phone/Samsung mobile phone/difference "; Introduce synonym near synonym and generate corresponding retrieval type, " i Phone and Samsung mobile phone and (different or distinguishes or difference ornull) "; Then existing database is retrieved.Answer given thus, often just include existing entry or the article of above keyword and its near synonym, although these entries or article relate to " i Phone " and " Samsung mobile phone " simultaneously, or give also both " difference ", but search for " i Phone " and " the Samsung mobile phone " that can not be directed to user provide pointed comparison functionally, but user is needed to find answer targetedly by reading entry or article.
By comparison, the nan-machine interrogation's method given by the present invention, then can provide answer more targetedly for the problem of user's input, provide explanation below with reference to accompanying drawing.Fig. 1 is the exemplary plot of the overall process process of nan-machine interrogation's method according to knowledge based collection of illustrative plates of the present invention.
According to the overall process process of nan-machine interrogation's method as shown in Figure 1, first machine receives the statement " it is different what i Phone and Samsung mobile phone have " inputted by user; And then machine is to statement through word segmentation processing, become " i Phone/with/Samsung mobile phone/have/what/different "; Then entity extraction is carried out to word, draw " i Phone: noun, apple, mobile phone/Samsung mobile phone: noun, apple, mobile phone ", and " difference " relation belonging to word, correspond to " comparison ".Different from the simple retrieval database of prior art, in the present invention, machine can load knowledge mapping, and after successfully loading knowledge mapping, using " i Phone " and " Samsung mobile phone " these two knowledge points as knowledge entrance, knowledge reasoning is carried out, to carry out entity lookup by recurrence reasoning.If when failing to find the knowledge point corresponding with " i Phone " and " Samsung mobile phone " in knowledge mapping, then reasoning failure.After find the functional attributes corresponding to knowledge point " i Phone " and " Samsung mobile phone " in knowledge mapping, reasoning success, and machine will get the result of its differentiation to be supplied to user as answer.According to nan-machine interrogation's method of the present invention, for the situation of the answer given by the problems referred to above compared to prior art, no longer need the user that puts question to by reading a large amount of relevant entry or answer found in article oneself, but more directly can provide and have more answer targetedly.
Below with reference to accompanying drawing, specifically describe the concrete treatment scheme according to nan-machine interrogation's method of the present invention.Fig. 2 is the example flow diagram of nan-machine interrogation's method of knowledge based collection of illustrative plates according to an embodiment of the invention.As shown in Figure 2, this nan-machine interrogation's method comprises step: ST10, receives the statement of user's input and carries out participle to the statement of user's input; ST20, carries out entity extraction to the word obtained by described participle; ST30, utilizes knowledge mapping to carry out knowledge reasoning to extracting by described entity the entity information obtained; And ST40, the result according to described knowledge reasoning provides feedback to the statement that described user inputs.
Carry out participle technique to the statement of input and belong to natural language processing technique field, for a statement, by the knowledge of oneself, the mankind can judge which is notional word, which is function word, and then understands its implication, then need to carry out participle for machine processing.In the step ST10 of participle, by segmentation methods, the statement that user inputs is cut, retain the stop words of relation belonging to Feature Words, and remove remaining stop words and redundant information.In the present invention, special restriction is not done for segmentation methods, the existing segmenting method based on string matching, the various concrete segmentation methods of segmenting method based on the segmenting method understood and Corpus--based Method can be adopted.
Specifically, for " whom the author of master of slamdunking is ", can become after participle " master of slamdunking// author/be/who ".In order to save storage space and improve treatment effeciency, after participle, the word that those practical significances can be gone further little or word, namely retain the stop words of relation belonging to Feature Words, and remove remaining stop words and redundant information.For above-mentioned example statement, wherein " " and "Yes" belong to stop words, " " then belongs to redundant information as auxiliary words of mood, although wherein "Yes" was a stop words originally, but here it has the meaning of relationships point, so can be retained, after removal stop words and redundant information, statement becomes " master of slamdunking/author/be/who ".
For stop words, provide on the basis of available " vocabulary of stopping using " in the prior art, the present invention is also provided with further " relation stop using vocabulary ", the stop words of a series of relation belonging to Feature Words is stored in " relation stop using vocabulary ", such as such as, "Yes", " having ", " being " etc.When carrying out removing the process of stop words, by searching " relation stop using vocabulary " and retaining the stop words of relation belonging to Feature Words, fine granularity and the accuracy of word can be ensured.
In the step ST20 that described entity extracts, utilize the entity attribute of the word obtained to mark it, wherein, described entity attribute comprises the part of speech of institute's predicate, the dictionary lexical or textual analysis of institute's predicate or the product classification corresponding to institute's predicate.In the present invention, for entity extract algorithm do not do special restriction, existing various concrete segmentation methods can be adopted, as long as it can realize carrying out entity replacement to word, judge which be name, which be commodity, which be numeral etc. base conditioning.Preferably, in the present invention, utilize condition random field to carry out entity extraction, to comprise the entity replacement of Entity recognition, part-of-speech tagging etc. to word.
Still for " whom the author of master of slamdunking is ", become after word segmentation processing " master of slamdunking/author/be/who ", when entity extracts, because " master of slamdunking " one word both corresponded to as the books series products of caricature, also the audio-visual series products as animation is corresponded to, " master of slamdunking " can be labeled as caricature and animation, and due to determiner not clear and definite in the statement of above-mentioned input, so the result of these 2 kinds of entity marks can be retained; " author " one word as the title of the attribute of " master of slamdunking " own, can't be extracted, but relative can be used as in reasoning; " who " is personal pronoun, can be extracted and mark with " pronoun, personage "; Thus, for above-mentioned example, its entity extract result be " master of slamdunking: caricature/animation, who: pronoun.For the situation including determiner in statement, such as " whom the author of caricature master of slamdunking is ", after entity extracts, because include the modification of " caricature " this entity in statement, so only can retain " master of slamdunking " attribute as caricature, therefore " master of slamdunking " can be labeled as caricature, and the entity of this statement extract result be correspondingly " master of slamdunking: caricature; who: pronoun, personage ".
After completing entity and extracting, be different from the simple mapping table in prior art between knowledge point, the present invention carries out knowledge reasoning by utilizing knowledge mapping to obtained entity information.Existing available knowledge mapping in prior art, the Freebase etc. of such as Google.In the present invention, establish the knowledge mapping of the relation between knowledge point and knowledge point, the product preferably provided using website or APP or service are as knowledge point to set up knowledge mapping.Described knowledge mapping comprises the relation between knowledge point and knowledge point, wherein each described knowledge point is provided with independently in-degree point and out-degree point, and the relation between described knowledge point is the knowledge relation by setting up the classification between described knowledge point according to identical in-degree point and out-degree point, sets up knowledge non-directed graph and set up between described knowledge point.
In order to the facility illustrated, as shown in Figure 3, the graph of a relation of the example of knowledge mapping is given for football, wherein " physical culture ", " football ", " Mei Xi ", " Barcelona ", " Real Madrid ", " Mei Xi ", " sieve C ", " Spain " are knowledge points, and these knowledge points form knowledge non-directed graph by the relatival connection such as " motion ", " football player ", " football club ", " effect ", " position/subordinate ".
When setting up knowledge mapping, can mark the entity attribute of each knowledge point wherein, " master of slamdunking " in such as previous examples, its entity attribute can comprise caricature, animation, delivers the time, author, producer, price etc.; " Mei Xi " for another example in follow-up example, its attribute has personage, football player, sex, effect club etc., so that machine can identify, thus can draw desired content information after going to search knowledge mapping by entity.
For the knowledge mapping set up, it is not a kind of speech network of simple static state, preferably, can modify to the entity attribute of wherein knowledge point and supplement.Such as, " Mei Xi " in above-mentioned example, when setting up knowledge mapping, its " effect club " is labeled as " Barcelona " by meeting basis at that time situation, in order to machine still can provide correct information when the follow-up change along with situation, " effect club " is remained with to the option of amendment, and be the attribute of " man " etc. these not malleables usually for " football player " and " sex ", then need not retain the option revised, to reduce system complexity for it.In addition, preferably, for the knowledge mapping set up, can carry out newly-increased for knowledge point and the entity attribute of wherein knowledge point be supplemented.Such as, " Wuwei " word, iff place name, tourism/ticketing service to its entity attribute labeling based on current provided GT grand touring product or service, when electric commercial business industry adds books or animation product and can provide the product about " Yu Yu Hakusho ", even if when user have input the statement comprising " Wuwei " and " caricature ", but because knowledge reasoning correctly can not obtain the information about " Yu Yu Hakusho ", and correct result can not be provided, and then the conclusion of the business chance of businesses may be lost.In the supplementing of the newly-increased of knowledge point and entity attribute, in order to reduce the complexity of system and reduce storage load, be not that corresponding knowledge point or entity attribute are all increased to whole neologisms, preferably, newly-increased knowledge point or supplementary entity attribute is come based on product newly-increased in website or APP or service.
In knowledge mapping of the present invention, there are independently in-degree point and out-degree point in each knowledge point, enters read point and out-degree point and sets up one-level, secondary, the knowledge relation of three grades, finally set up a knowledge non-directed graph according to identical.Referring again to knowledge mapping illustrated in fig. 3 is example, wherein " physical culture " is the knowledge point of one-level, " football " is the knowledge point of secondary, " Mei Xi ", " Barcelona ", " sieve C ", " Real Madrid " are the knowledge of three grades, the knowledge of three grades has allowed other high level knowledge to do in-degree point, such as, " Mei Xi " is personage, and " Barcelona " is club etc.Based on the knowledge mapping of example, can infer him by " Mei Xi " is the personage in " physical culture " field or the personage in " football " field, " Mei Xi " can be inferred by the relation between peer to serve " Barcelona ", and " Mei Xi " and " Real Madrid " current not common factor etc.
In knowledge mapping of the present invention, visit this knowledge point by the in-degree point of the different stage of each knowledge point, visit knowledge point at the same level at the out-degree point by this knowledge point.Line between knowledge point embodies the relation between each knowledge point, and such as " Li Yuan " can be connected by " father and son " relation with " Li Shih-min ", and " master of slamdunking " can be connected by " author " relation with " Takehiko Inoue ".
Utilizing knowledge mapping to carry out in the step ST30 of knowledge reasoning to extracting by described entity the entity information obtained, preferably, when comprising an entity information, in described knowledge mapping, search the knowledge point corresponding to described entity information; Acquisition described knowledge point after, terminate the step of described reasoning, and provide and the content information corresponding to described knowledge point, and in order to can quick-searching to described knowledge point, provide the positional information of described knowledge point in described knowledge mapping; And when not finding described knowledge point when traveling through described knowledge mapping, terminating the step of described reasoning, and providing the feedback without accordingly result.
Still for knowledge mapping illustrated in fig. 3, the statement of such as user's input is " whom Mei Xi is ", can become after word segmentation processing " Mei Xi/be/who ", entity can become after extracting process " Mei Xi: personage, football player, who: pronoun, personage ".When knowledge reasoning, find the knowledge point corresponding to entity information " Mei Xi " in knowledge mapping, and provide the information " football player " corresponding to " personage ", reasoning terminates.The statement of such as user's input is " whom Marcos Soares is " again, can become after word segmentation processing " Marcos Soares/be/who ", cannot find the knowledge point corresponding to entity information " Marcos Soares " in current knowledge collection of illustrative plates, therefore reasoning terminates, and provides the feedback of " without accordingly result ".
Carry out in the step ST30 of knowledge reasoning utilizing knowledge mapping to extracting by described entity the entity information obtained, preferably, when comprising many entity informations, in described knowledge mapping, a) search the first knowledge point corresponding in described entity information; B) by the out-degree point of described first knowledge point, utilize described first knowledge point and correspond to the relation between another the second knowledge point in described entity information, in described knowledge mapping, search described second knowledge point; C) repeat above-mentioned b step, search until completed for the whole knowledge point corresponded in described entity information, terminate the step of described reasoning, and provide and the content information corresponding to described knowledge point; And d) repeat above-mentioned b step, when not finding the knowledge point that will search when traveling through described knowledge mapping, terminating the step of described reasoning, and providing the feedback without accordingly result.
Still for knowledge mapping illustrated in fig. 3, such as the statement of user's input is " club at Mei Xi place is ", can become after word segmentation processing " Mei Xi/place/club/be ", entity can become " Mei Xi: personage, football player, club: football, group, place " after extracting process.When knowledge reasoning, the knowledge point corresponding to entity information " Mei Xi " is found in knowledge mapping, by the point using " Mei Xi " as out-degree, using " effect " as relative, " Barcelona " this knowledge point can be found as in-degree point, obtain thus " club at Mei Xi place is Barcelona ", reasoning terminates thus.For the example of aforesaid " whom author of master of slamdunking is ", when knowledge reasoning, the knowledge point corresponding to entity information " master of slamdunking " is found in knowledge mapping, by the point using " master of slamdunking " as out-degree, to find in the knowledge point being marked as " personage " close with it be the corresponding knowledge point of " author " as in-degree point, reasoning terminates thus.
For problem " club at Mei Xi place is ", inventor inputs the problems referred to above to make comparisons with method of the present invention at the existing search engine page of knowledge mapping technology that do not adopt.The result that existing search engine provides includes the statement found in existing database by full character match, which includes Similar Problems such as " which club at Mei Xi place are ", and answer still needs user by consulting corresponding web page to find, and also include in result about the information corresponding to " Mei Xi " this entry, and concrete specific aim answer still needs user to pass through to read corresponding word to find and obtain.By comparison, be then give concrete answer to enquirement according to method of the present invention, this is for can be more directly perceived user.
The statement of such as user's input is " club at Marcos Soares place is " again, can become after word segmentation processing " Marcos Soares/place/club/be ", the knowledge point corresponding to entity information " Marcos Soares " cannot be found in the knowledge mapping of example shown in current Fig. 3, therefore reasoning terminates, and provides the feedback of " without accordingly result ".
For another example, for " why expensive than common keyboard mechanical keyboard is ", inventor have input the problems referred to above to make comparisons with method of the present invention in the existing interaction page of nan-machine interrogation's system of knowledge mapping technology that do not adopt.The result that existing nan-machine interrogation's system provides is such as " because supplier is different; and the cause influence such as different businessmans sales promotion; commodity price may exist difference ", this is only aimed at " expensive " gives general pervasive explanation, and does not provide for two comparison objects of user's input and anyly to compare targetedly.
For the problems referred to above " why expensive than common keyboard mechanical keyboard is " of user's input, first word segmentation processing can be carried out to it according to method of the present invention, can become after word segmentation processing " mechanical keyboard/why/ratio/common keyboard/expensive ", entity can become " mechanical keyboard: noun, input product, common keyboard: noun, input product " after extracting process.There is the relatival key word of conduct of " expensive " and " ratio " here, when knowledge reasoning, the knowledge point corresponding to entity information " mechanical keyboard " and " common keyboard/membrane keyboard " can be found in knowledge mapping, do attribute compare by carrying out difference comparsion to the cost of the attribute belonging to both, and the comparative result obtained about the materials, serviceable life, Consumer's Experience etc. of such as product is used as answer.
Fig. 4 is the example block diagram of the nan-machine interrogation's system according to knowledge based collection of illustrative plates of the present invention, and the nan-machine interrogation's system 100 shown in it comprises: load module 10, for receiving the statement of user's input; Word-dividing mode 20, carries out participle for the statement inputted user; Entity abstraction module 30, for carrying out entity extraction to the word obtained by described participle; Knowledge mapping module 40, for storing the knowledge mapping of the relation comprised between knowledge point and knowledge point; Knowledge reasoning module 50, carries out knowledge reasoning for utilizing described knowledge mapping to extracting by described entity the entity information obtained; And output module 60, for the result according to described knowledge reasoning, feedback is provided to the statement that described user inputs.
Preferably, in one embodiment, load module 10 can receive with the statement of the user of words input, voice typing and/or additive method typing known in the art input.
Preferably, in one embodiment, word-dividing mode 20 is cut the statement that user inputs by segmentation methods, retains the stop words of relation belonging to Feature Words, and removes remaining stop words and redundant information.
Preferably, in one embodiment, entity abstraction module 30 utilizes the entity attribute of the word obtained to mark it, and wherein, described entity attribute comprises the part of speech of institute's predicate, the dictionary lexical or textual analysis of institute's predicate or the product classification corresponding to institute's predicate.
Preferably, in one embodiment, knowledge mapping module 40 is by arranging independently in-degree point and out-degree point for each described knowledge point, the knowledge relation of the classification between described knowledge point is set up according to identical in-degree point and out-degree point, between described knowledge point, set up knowledge non-directed graph, store the relation between the knowledge point of described knowledge mapping and knowledge point.
Preferably, in one embodiment, knowledge reasoning module 50, when obtaining the knowledge point corresponding to described entity information by searching in described knowledge mapping, provides the positional information in described knowledge mapping with the content information corresponding to described knowledge point and described knowledge point; And when not finding described knowledge point when traveling through described knowledge mapping, provide the feedback without accordingly result.
Preferably, in one embodiment, knowledge reasoning module 50 searches the first knowledge point corresponding in described entity information in described knowledge mapping; By the out-degree point of described first knowledge point, utilize described first knowledge point and correspond to the relation between another the second knowledge point in described entity information, in described knowledge mapping, search described second knowledge point; When repetition above-mentioned steps, until for the whole knowledge point corresponded in described entity information completed search time, provide and the content information corresponding to described knowledge point; And when not finding when traveling through described knowledge mapping the knowledge point that will search, provide the feedback without accordingly result.
Preferably, in one embodiment, output module 60 can in a visual manner, mode and/or other modes known in the art can be listened to provide respective feedback to user's read statement.
Nan-machine interrogation's method and system of the present invention as above, can be applicable to the automatic answering system as JIMI, also can be embedded in search engine, the knowledge reasoning of knowledge based collection of illustrative plates is carried out with the problem inputted user, the depth & wideth of knowledge reasoning is significantly improved, and then provides answer more targetedly as feedback.
Below ultimate principle of the present invention is described in conjunction with specific embodiments, but, it is to be noted, for those of ordinary skill in the art, whole or any step or the parts of method and apparatus of the present invention can be understood, can in the network of any calculation element (comprising processor, storage medium etc.) or calculation element, realized with hardware, firmware, software or their combination, this is that those of ordinary skill in the art use their basic programming skill just can realize when having read explanation of the present invention.
Therefore, object of the present invention can also be realized by an operation program or batch processing on any calculation element.Described calculation element can be known fexible unit.Therefore, object of the present invention also can realize only by the program product of providing package containing the program code realizing described method or device.That is, such program product also forms the present invention, and the storage medium storing such program product also forms the present invention.Obviously, described storage medium can be any storage medium developed in any known storage medium or future.
Also it is pointed out that in apparatus and method of the present invention, obviously, each parts or each step can decompose and/or reconfigure.These decompose and/or reconfigure and should be considered as equivalents of the present invention.Further, the step performing above-mentioned series of processes can order naturally following the instructions perform in chronological order, but does not need necessarily to perform according to time sequencing.Some step can walk abreast or perform independently of one another.
Above-mentioned embodiment, does not form limiting the scope of the invention.It is to be understood that depend on designing requirement and other factors, various amendment, combination, sub-portfolio can be there is and substitute in those skilled in the art.Any amendment done within the spirit and principles in the present invention, equivalent replacement and improvement etc., all should be included within scope.
Claims (12)
1. nan-machine interrogation's method of knowledge based collection of illustrative plates, described method comprises:
Receive the statement of user's input and participle is carried out to described statement;
Entity extraction is carried out to the word obtained by described participle;
Knowledge mapping is utilized to carry out knowledge reasoning to extracting by described entity the entity information obtained; And
Result according to described knowledge reasoning provides feedback to the statement that described user inputs.
2. nan-machine interrogation's method according to claim 1, wherein, the step of described participle comprises: cut the statement that user inputs by segmentation methods, retains the stop words of relation belonging to Feature Words, and removes remaining stop words and redundant information.
3. nan-machine interrogation's method according to claim 1, wherein, the step that described entity extracts comprises: utilize the entity attribute of the word obtained to mark it, wherein, described entity attribute comprises the part of speech of institute's predicate, the dictionary lexical or textual analysis of institute's predicate or the product classification corresponding to institute's predicate.
4. nan-machine interrogation's method according to claim 1, wherein, described knowledge mapping comprises the relation between knowledge point and knowledge point, wherein each described knowledge point is provided with independently in-degree point and out-degree point, and the relation between described knowledge point is the knowledge relation by setting up the classification between described knowledge point according to identical in-degree point and out-degree point, sets up knowledge non-directed graph and set up between described knowledge point.
5. nan-machine interrogation's method according to claim 1, wherein, the step of described knowledge reasoning comprises:
The knowledge point corresponding to described entity information is searched in described knowledge mapping;
After the described knowledge point of acquisition, terminate the step of described reasoning, and provide the positional information in described knowledge mapping with the content information corresponding to described knowledge point and described knowledge point; And
When not finding described knowledge point when traveling through described knowledge mapping, terminating the step of described reasoning, and providing the feedback without accordingly result.
6. nan-machine interrogation's method according to claim 1, wherein, the step of described knowledge reasoning comprises:
A) in described knowledge mapping, search the first knowledge point corresponding in described entity information;
B) by the out-degree point of described first knowledge point, utilize described first knowledge point and correspond to the relation between another the second knowledge point in described entity information, in described knowledge mapping, search described second knowledge point;
C) repeat above-mentioned b step, search until completed for the whole knowledge point corresponded in described entity information, terminate the step of described reasoning, and provide and the content information corresponding to described knowledge point; And
D) repeat above-mentioned b step, when not finding the knowledge point that will search when traveling through described knowledge mapping, terminating the step of described reasoning, and providing the feedback without accordingly result.
7. nan-machine interrogation's system of knowledge based collection of illustrative plates, described system comprises:
Load module, for receiving the statement of user's input;
Word-dividing mode, carries out participle for the statement inputted user;
Entity abstraction module, for carrying out entity extraction to the word obtained by described participle;
Knowledge mapping module, for storing the knowledge mapping of the relation comprised between knowledge point and knowledge point;
Knowledge reasoning module, carries out knowledge reasoning for utilizing described knowledge mapping to extracting by described entity the entity information obtained; And
Output module, provides feedback for the result according to described knowledge reasoning to the statement that described user inputs.
8. nan-machine interrogation's system according to claim 7, wherein, described word-dividing mode is cut the statement that user inputs by segmentation methods, retains the stop words of relation belonging to Feature Words, and removes remaining stop words and redundant information.
9. nan-machine interrogation's system according to claim 7, wherein, described entity abstraction module utilizes the entity attribute of the word obtained to mark it, and wherein, described entity attribute comprises the part of speech of institute's predicate, the dictionary lexical or textual analysis of institute's predicate or the product classification corresponding to institute's predicate.
10. nan-machine interrogation's system according to claim 7, wherein, described knowledge mapping module is by arranging independently in-degree point and out-degree point for each described knowledge point, the knowledge relation of the classification between described knowledge point is set up according to identical in-degree point and out-degree point, between described knowledge point, set up knowledge non-directed graph, store the relation between the knowledge point of described knowledge mapping and knowledge point.
11. nan-machine interrogation's systems according to claim 7, wherein, described knowledge reasoning module, when obtaining the knowledge point corresponding to described entity information by searching in described knowledge mapping, provides the positional information in described knowledge mapping with the content information corresponding to described knowledge point and described knowledge point; And when not finding described knowledge point when traveling through described knowledge mapping, provide the feedback without accordingly result.
12. nan-machine interrogation's systems according to claim 7, wherein, described knowledge reasoning module searches the first knowledge point corresponding in described entity information in described knowledge mapping; By the out-degree point of described first knowledge point, utilize described first knowledge point and correspond to the relation between another the second knowledge point in described entity information, in described knowledge mapping, search described second knowledge point; When repetition above-mentioned steps, until for the whole knowledge point corresponded in described entity information completed search time, provide and the content information corresponding to described knowledge point; And when not finding when traveling through described knowledge mapping the knowledge point that will search, provide the feedback without accordingly result.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510383452.7A CN105095195B (en) | 2015-07-03 | 2015-07-03 | Nan-machine interrogation's method and system of knowledge based collection of illustrative plates |
HK16105720.8A HK1217789A1 (en) | 2015-07-03 | 2016-05-18 | Method and system for human-machine questioning and answering based on knowledge graph |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510383452.7A CN105095195B (en) | 2015-07-03 | 2015-07-03 | Nan-machine interrogation's method and system of knowledge based collection of illustrative plates |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105095195A true CN105095195A (en) | 2015-11-25 |
CN105095195B CN105095195B (en) | 2018-09-18 |
Family
ID=54575666
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510383452.7A Active CN105095195B (en) | 2015-07-03 | 2015-07-03 | Nan-machine interrogation's method and system of knowledge based collection of illustrative plates |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN105095195B (en) |
HK (1) | HK1217789A1 (en) |
Cited By (51)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105677822A (en) * | 2016-01-05 | 2016-06-15 | 首都师范大学 | Enrollment automatic question-answering method and system based on conversation robot |
CN105843875A (en) * | 2016-03-18 | 2016-08-10 | 北京光年无限科技有限公司 | Smart robot-oriented question and answer data processing method and apparatus |
CN106095932A (en) * | 2016-06-13 | 2016-11-09 | 竹间智能科技(上海)有限公司 | Encyclopaedic knowledge question sentence recognition methods and device |
CN106294325A (en) * | 2016-08-11 | 2017-01-04 | 海信集团有限公司 | The optimization method and device of spatial term statement |
CN106599179A (en) * | 2016-12-13 | 2017-04-26 | 竹间智能科技(上海)有限公司 | Method and device for controlling man-machine conversation based on fusion of knowledge graph and memory graph |
CN106934012A (en) * | 2017-03-10 | 2017-07-07 | 上海数眼科技发展有限公司 | A kind of question answering in natural language method and system of knowledge based collection of illustrative plates |
CN107180059A (en) * | 2016-03-11 | 2017-09-19 | 北大方正集团有限公司 | Data retrieval method and data retrieval system |
CN107391512A (en) * | 2016-05-17 | 2017-11-24 | 北京邮电大学 | The method and apparatus of knowledge mapping prediction |
CN107545000A (en) * | 2016-06-28 | 2018-01-05 | 百度在线网络技术(北京)有限公司 | The information-pushing method and device of knowledge based collection of illustrative plates |
CN107679039A (en) * | 2017-10-17 | 2018-02-09 | 北京百度网讯科技有限公司 | The method and apparatus being intended to for determining sentence |
CN107748757A (en) * | 2017-09-21 | 2018-03-02 | 北京航空航天大学 | A kind of answering method of knowledge based collection of illustrative plates |
CN107766483A (en) * | 2017-10-13 | 2018-03-06 | 华中科技大学 | The interactive answering method and system of a kind of knowledge based collection of illustrative plates |
CN107885842A (en) * | 2017-11-10 | 2018-04-06 | 上海智臻智能网络科技股份有限公司 | Method, apparatus, server and the storage medium of intelligent answer |
CN107943998A (en) * | 2017-12-05 | 2018-04-20 | 竹间智能科技(上海)有限公司 | A kind of human-machine conversation control system and method for knowledge based collection of illustrative plates |
CN107992528A (en) * | 2017-11-13 | 2018-05-04 | 清华大学 | Utilize more relation question answering systems of interpretable inference network |
CN108108982A (en) * | 2016-11-25 | 2018-06-01 | 百度在线网络技术(北京)有限公司 | A kind of recognition methods of authorization message and device |
CN108452526A (en) * | 2017-11-28 | 2018-08-28 | 腾讯科技(上海)有限公司 | The querying method and device, storage medium, electronic device for failure cause of playing |
CN108509563A (en) * | 2018-03-23 | 2018-09-07 | 深圳狗尾草智能科技有限公司 | Robot reasoning association method, device, equipment and the medium of knowledge based collection of illustrative plates |
CN108519998A (en) * | 2018-03-07 | 2018-09-11 | 北京云知声信息技术有限公司 | The problem of knowledge based collection of illustrative plates bootstrap technique and device |
CN108733654A (en) * | 2018-05-21 | 2018-11-02 | 宁波薄言信息技术有限公司 | A kind of information processing method |
CN109062896A (en) * | 2018-07-25 | 2018-12-21 | 南京瓦尔基里网络科技有限公司 | A kind of matching process and system based on artificial intelligence words art model |
WO2019000170A1 (en) * | 2017-06-26 | 2019-01-03 | Microsoft Technology Licensing, Llc | Generating responses in automated chatting |
CN109145102A (en) * | 2018-09-06 | 2019-01-04 | 杭州安恒信息技术股份有限公司 | Intelligent answer method and its knowledge mapping system constituting method, device, equipment |
CN109271504A (en) * | 2018-11-07 | 2019-01-25 | 爱因互动科技发展(北京)有限公司 | The method of the reasoning dialogue of knowledge based map |
CN109300472A (en) * | 2018-12-21 | 2019-02-01 | 深圳创维-Rgb电子有限公司 | A kind of audio recognition method, device, equipment and medium |
CN109388793A (en) * | 2017-08-03 | 2019-02-26 | 阿里巴巴集团控股有限公司 | Entity mask method, intension recognizing method and corresponding intrument, computer storage medium |
CN109446387A (en) * | 2018-10-09 | 2019-03-08 | 众蚁(上海)信息技术有限公司 | A kind of Owners Committee's intelligent Answer System based on artificial intelligence |
CN109492077A (en) * | 2018-09-29 | 2019-03-19 | 北明智通(北京)科技有限公司 | The petrochemical field answering method and system of knowledge based map |
CN109522465A (en) * | 2018-10-22 | 2019-03-26 | 国家电网公司 | The semantic searching method and device of knowledge based map |
CN109543007A (en) * | 2018-10-16 | 2019-03-29 | 深圳壹账通智能科技有限公司 | Put question to data creation method, device, computer equipment and storage medium |
CN109597894A (en) * | 2018-09-30 | 2019-04-09 | 阿里巴巴集团控股有限公司 | A kind of correlation model generation method and device, a kind of data correlation method and device |
CN109844741A (en) * | 2017-06-29 | 2019-06-04 | 微软技术许可有限责任公司 | Response is generated in automatic chatting |
CN109902165A (en) * | 2019-03-08 | 2019-06-18 | 中国科学院自动化研究所 | Intelligent interactive answering method, system, device based on Markov Logic Networks |
CN109933671A (en) * | 2019-01-31 | 2019-06-25 | 平安科技(深圳)有限公司 | Construct method, apparatus, computer equipment and the storage medium of personal knowledge map |
CN110019710A (en) * | 2017-11-27 | 2019-07-16 | 厦门快商通信息技术有限公司 | A kind of topic forest formula interactive method and system |
CN110175227A (en) * | 2019-05-10 | 2019-08-27 | 神思电子技术股份有限公司 | A kind of dialogue auxiliary system based on form a team study and level reasoning |
US10423726B2 (en) | 2018-01-10 | 2019-09-24 | International Business Machines Corporation | Machine learning to integrate knowledge and natural language processing |
US10606958B2 (en) | 2018-01-10 | 2020-03-31 | International Business Machines Corporation | Machine learning modification and natural language processing |
CN111095259A (en) * | 2017-10-25 | 2020-05-01 | 谷歌有限责任公司 | Natural language processing using N-GRAM machines |
CN111291168A (en) * | 2018-12-07 | 2020-06-16 | 北大方正集团有限公司 | Book retrieval method and device and readable storage medium |
CN111460172A (en) * | 2020-03-31 | 2020-07-28 | 北京小米移动软件有限公司 | Method and device for determining answers to product questions and electronic equipment |
CN111488741A (en) * | 2020-04-14 | 2020-08-04 | 税友软件集团股份有限公司 | Tax knowledge data semantic annotation method and related device |
CN111552880A (en) * | 2020-04-30 | 2020-08-18 | 杭州网易再顾科技有限公司 | Data processing method, device and medium based on knowledge graph and electronic equipment |
US10776586B2 (en) | 2018-01-10 | 2020-09-15 | International Business Machines Corporation | Machine learning to integrate knowledge and augment natural language processing |
CN111859974A (en) * | 2019-04-22 | 2020-10-30 | 广东小天才科技有限公司 | Semantic disambiguation method and device combined with knowledge graph and intelligent learning equipment |
CN111930916A (en) * | 2020-09-18 | 2020-11-13 | 北京百度网讯科技有限公司 | Dialog generation method and device, electronic equipment and storage medium |
CN111949855A (en) * | 2020-07-31 | 2020-11-17 | 国网上海市电力公司 | Knowledge map-based engineering technology knowledge retrieval platform and method thereof |
CN111966834A (en) * | 2020-07-29 | 2020-11-20 | 深圳市元征科技股份有限公司 | File generation method, file generation device and server |
CN112231445A (en) * | 2020-03-27 | 2021-01-15 | 北京来也网络科技有限公司 | Searching method, device, equipment and storage medium combining RPA and AI |
CN113434658A (en) * | 2021-08-25 | 2021-09-24 | 西安热工研究院有限公司 | Thermal power generating unit operation question-answer generation method, system, equipment and readable storage medium |
CN117573849A (en) * | 2024-01-16 | 2024-02-20 | 之江实验室 | Knowledge graph multi-hop question-answering method, device, equipment and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103902652A (en) * | 2014-02-27 | 2014-07-02 | 深圳市智搜信息技术有限公司 | Automatic question-answering system |
CN104252533A (en) * | 2014-09-12 | 2014-12-31 | 百度在线网络技术(北京)有限公司 | Search method and search device |
CN104462504A (en) * | 2014-12-19 | 2015-03-25 | 北京奇虎科技有限公司 | Method and device for providing reasoning process data in search |
-
2015
- 2015-07-03 CN CN201510383452.7A patent/CN105095195B/en active Active
-
2016
- 2016-05-18 HK HK16105720.8A patent/HK1217789A1/en unknown
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103902652A (en) * | 2014-02-27 | 2014-07-02 | 深圳市智搜信息技术有限公司 | Automatic question-answering system |
CN104252533A (en) * | 2014-09-12 | 2014-12-31 | 百度在线网络技术(北京)有限公司 | Search method and search device |
CN104462504A (en) * | 2014-12-19 | 2015-03-25 | 北京奇虎科技有限公司 | Method and device for providing reasoning process data in search |
Non-Patent Citations (2)
Title |
---|
PAWEL JURCZYK 等: "Discovering Authorities in Question Answer Communities by Using Link Analysis", 《ACM》 * |
曾庆鹏 等: "网络教学平台中问答系统的关键技术研究", 《计算机与现代化》 * |
Cited By (77)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105677822A (en) * | 2016-01-05 | 2016-06-15 | 首都师范大学 | Enrollment automatic question-answering method and system based on conversation robot |
CN107180059A (en) * | 2016-03-11 | 2017-09-19 | 北大方正集团有限公司 | Data retrieval method and data retrieval system |
CN105843875B (en) * | 2016-03-18 | 2019-09-13 | 北京光年无限科技有限公司 | A kind of question and answer data processing method and device towards intelligent robot |
CN105843875A (en) * | 2016-03-18 | 2016-08-10 | 北京光年无限科技有限公司 | Smart robot-oriented question and answer data processing method and apparatus |
CN107391512A (en) * | 2016-05-17 | 2017-11-24 | 北京邮电大学 | The method and apparatus of knowledge mapping prediction |
CN107391512B (en) * | 2016-05-17 | 2021-05-11 | 北京邮电大学 | Method and device for predicting knowledge graph |
CN106095932A (en) * | 2016-06-13 | 2016-11-09 | 竹间智能科技(上海)有限公司 | Encyclopaedic knowledge question sentence recognition methods and device |
CN107545000A (en) * | 2016-06-28 | 2018-01-05 | 百度在线网络技术(北京)有限公司 | The information-pushing method and device of knowledge based collection of illustrative plates |
CN106294325A (en) * | 2016-08-11 | 2017-01-04 | 海信集团有限公司 | The optimization method and device of spatial term statement |
CN106294325B (en) * | 2016-08-11 | 2019-01-04 | 海信集团有限公司 | The optimization method and device of spatial term sentence |
CN108108982A (en) * | 2016-11-25 | 2018-06-01 | 百度在线网络技术(北京)有限公司 | A kind of recognition methods of authorization message and device |
CN106599179A (en) * | 2016-12-13 | 2017-04-26 | 竹间智能科技(上海)有限公司 | Method and device for controlling man-machine conversation based on fusion of knowledge graph and memory graph |
CN106599179B (en) * | 2016-12-13 | 2020-02-14 | 竹间智能科技(上海)有限公司 | Man-machine conversation control method and device integrating knowledge graph and memory graph |
CN106934012B (en) * | 2017-03-10 | 2020-05-08 | 上海数眼科技发展有限公司 | Natural language question-answering implementation method and system based on knowledge graph |
CN106934012A (en) * | 2017-03-10 | 2017-07-07 | 上海数眼科技发展有限公司 | A kind of question answering in natural language method and system of knowledge based collection of illustrative plates |
CN109844743A (en) * | 2017-06-26 | 2019-06-04 | 微软技术许可有限责任公司 | Response is generated in automatic chatting |
US11586810B2 (en) | 2017-06-26 | 2023-02-21 | Microsoft Technology Licensing, Llc | Generating responses in automated chatting |
CN109844743B (en) * | 2017-06-26 | 2023-10-17 | 微软技术许可有限责任公司 | Generating responses in automated chat |
WO2019000170A1 (en) * | 2017-06-26 | 2019-01-03 | Microsoft Technology Licensing, Llc | Generating responses in automated chatting |
CN109844741A (en) * | 2017-06-29 | 2019-06-04 | 微软技术许可有限责任公司 | Response is generated in automatic chatting |
CN109844741B (en) * | 2017-06-29 | 2023-09-01 | 微软技术许可有限责任公司 | Generating responses in automated chat |
CN109388793A (en) * | 2017-08-03 | 2019-02-26 | 阿里巴巴集团控股有限公司 | Entity mask method, intension recognizing method and corresponding intrument, computer storage medium |
CN109388793B (en) * | 2017-08-03 | 2023-04-07 | 阿里巴巴集团控股有限公司 | Entity marking method, intention identification method, corresponding device and computer storage medium |
CN107748757A (en) * | 2017-09-21 | 2018-03-02 | 北京航空航天大学 | A kind of answering method of knowledge based collection of illustrative plates |
CN107748757B (en) * | 2017-09-21 | 2021-05-07 | 北京航空航天大学 | Question-answering method based on knowledge graph |
CN107766483A (en) * | 2017-10-13 | 2018-03-06 | 华中科技大学 | The interactive answering method and system of a kind of knowledge based collection of illustrative plates |
CN107679039A (en) * | 2017-10-17 | 2018-02-09 | 北京百度网讯科技有限公司 | The method and apparatus being intended to for determining sentence |
CN107679039B (en) * | 2017-10-17 | 2020-12-29 | 北京百度网讯科技有限公司 | Method and device for determining statement intention |
CN111095259B (en) * | 2017-10-25 | 2023-08-22 | 谷歌有限责任公司 | Natural Language Processing Using N-GRAM Machines |
CN111095259A (en) * | 2017-10-25 | 2020-05-01 | 谷歌有限责任公司 | Natural language processing using N-GRAM machines |
US11947917B2 (en) | 2017-10-25 | 2024-04-02 | Google Llc | Natural language processing with an n-gram machine |
CN107885842A (en) * | 2017-11-10 | 2018-04-06 | 上海智臻智能网络科技股份有限公司 | Method, apparatus, server and the storage medium of intelligent answer |
CN107992528A (en) * | 2017-11-13 | 2018-05-04 | 清华大学 | Utilize more relation question answering systems of interpretable inference network |
CN107992528B (en) * | 2017-11-13 | 2022-07-05 | 清华大学 | Multi-relational question-answering system using interpretable reasoning network |
CN110019710A (en) * | 2017-11-27 | 2019-07-16 | 厦门快商通信息技术有限公司 | A kind of topic forest formula interactive method and system |
CN108452526B (en) * | 2017-11-28 | 2020-12-25 | 腾讯科技(上海)有限公司 | Game fault reason query method and device, storage medium and electronic device |
CN108452526A (en) * | 2017-11-28 | 2018-08-28 | 腾讯科技(上海)有限公司 | The querying method and device, storage medium, electronic device for failure cause of playing |
CN107943998A (en) * | 2017-12-05 | 2018-04-20 | 竹间智能科技(上海)有限公司 | A kind of human-machine conversation control system and method for knowledge based collection of illustrative plates |
US10423726B2 (en) | 2018-01-10 | 2019-09-24 | International Business Machines Corporation | Machine learning to integrate knowledge and natural language processing |
US10599780B2 (en) | 2018-01-10 | 2020-03-24 | International Business Machines Corporation | Machine learning to integrate knowledge and natural language processing |
US10606958B2 (en) | 2018-01-10 | 2020-03-31 | International Business Machines Corporation | Machine learning modification and natural language processing |
US10776586B2 (en) | 2018-01-10 | 2020-09-15 | International Business Machines Corporation | Machine learning to integrate knowledge and augment natural language processing |
US10846485B2 (en) | 2018-01-10 | 2020-11-24 | International Business Machines Corporation | Machine learning model modification and natural language processing |
CN108519998A (en) * | 2018-03-07 | 2018-09-11 | 北京云知声信息技术有限公司 | The problem of knowledge based collection of illustrative plates bootstrap technique and device |
CN108519998B (en) * | 2018-03-07 | 2021-05-14 | 云知声智能科技股份有限公司 | Problem guiding method and device based on knowledge graph |
CN108509563A (en) * | 2018-03-23 | 2018-09-07 | 深圳狗尾草智能科技有限公司 | Robot reasoning association method, device, equipment and the medium of knowledge based collection of illustrative plates |
CN108733654A (en) * | 2018-05-21 | 2018-11-02 | 宁波薄言信息技术有限公司 | A kind of information processing method |
CN109062896A (en) * | 2018-07-25 | 2018-12-21 | 南京瓦尔基里网络科技有限公司 | A kind of matching process and system based on artificial intelligence words art model |
CN109145102B (en) * | 2018-09-06 | 2021-02-09 | 杭州安恒信息技术股份有限公司 | Intelligent question answering method and knowledge graph system construction method, device and equipment thereof |
CN109145102A (en) * | 2018-09-06 | 2019-01-04 | 杭州安恒信息技术股份有限公司 | Intelligent answer method and its knowledge mapping system constituting method, device, equipment |
CN109492077A (en) * | 2018-09-29 | 2019-03-19 | 北明智通(北京)科技有限公司 | The petrochemical field answering method and system of knowledge based map |
CN109597894B (en) * | 2018-09-30 | 2023-10-03 | 创新先进技术有限公司 | Correlation model generation method and device, and data correlation method and device |
CN109597894A (en) * | 2018-09-30 | 2019-04-09 | 阿里巴巴集团控股有限公司 | A kind of correlation model generation method and device, a kind of data correlation method and device |
CN109446387A (en) * | 2018-10-09 | 2019-03-08 | 众蚁(上海)信息技术有限公司 | A kind of Owners Committee's intelligent Answer System based on artificial intelligence |
CN109543007A (en) * | 2018-10-16 | 2019-03-29 | 深圳壹账通智能科技有限公司 | Put question to data creation method, device, computer equipment and storage medium |
CN109522465A (en) * | 2018-10-22 | 2019-03-26 | 国家电网公司 | The semantic searching method and device of knowledge based map |
CN109271504A (en) * | 2018-11-07 | 2019-01-25 | 爱因互动科技发展(北京)有限公司 | The method of the reasoning dialogue of knowledge based map |
CN109271504B (en) * | 2018-11-07 | 2021-06-25 | 爱因互动科技发展(北京)有限公司 | Inference dialogue method based on knowledge graph |
CN111291168A (en) * | 2018-12-07 | 2020-06-16 | 北大方正集团有限公司 | Book retrieval method and device and readable storage medium |
CN109300472A (en) * | 2018-12-21 | 2019-02-01 | 深圳创维-Rgb电子有限公司 | A kind of audio recognition method, device, equipment and medium |
CN109933671A (en) * | 2019-01-31 | 2019-06-25 | 平安科技(深圳)有限公司 | Construct method, apparatus, computer equipment and the storage medium of personal knowledge map |
CN109902165B (en) * | 2019-03-08 | 2021-02-23 | 中国科学院自动化研究所 | Intelligent interactive question-answering method, system and device based on Markov logic network |
CN109902165A (en) * | 2019-03-08 | 2019-06-18 | 中国科学院自动化研究所 | Intelligent interactive answering method, system, device based on Markov Logic Networks |
CN111859974A (en) * | 2019-04-22 | 2020-10-30 | 广东小天才科技有限公司 | Semantic disambiguation method and device combined with knowledge graph and intelligent learning equipment |
CN110175227B (en) * | 2019-05-10 | 2021-03-02 | 神思电子技术股份有限公司 | Dialogue auxiliary system based on team learning and hierarchical reasoning |
CN110175227A (en) * | 2019-05-10 | 2019-08-27 | 神思电子技术股份有限公司 | A kind of dialogue auxiliary system based on form a team study and level reasoning |
CN112231445A (en) * | 2020-03-27 | 2021-01-15 | 北京来也网络科技有限公司 | Searching method, device, equipment and storage medium combining RPA and AI |
CN111460172A (en) * | 2020-03-31 | 2020-07-28 | 北京小米移动软件有限公司 | Method and device for determining answers to product questions and electronic equipment |
CN111488741A (en) * | 2020-04-14 | 2020-08-04 | 税友软件集团股份有限公司 | Tax knowledge data semantic annotation method and related device |
CN111552880A (en) * | 2020-04-30 | 2020-08-18 | 杭州网易再顾科技有限公司 | Data processing method, device and medium based on knowledge graph and electronic equipment |
CN111966834A (en) * | 2020-07-29 | 2020-11-20 | 深圳市元征科技股份有限公司 | File generation method, file generation device and server |
CN111949855A (en) * | 2020-07-31 | 2020-11-17 | 国网上海市电力公司 | Knowledge map-based engineering technology knowledge retrieval platform and method thereof |
CN111930916A (en) * | 2020-09-18 | 2020-11-13 | 北京百度网讯科技有限公司 | Dialog generation method and device, electronic equipment and storage medium |
CN111930916B (en) * | 2020-09-18 | 2021-02-05 | 北京百度网讯科技有限公司 | Dialog generation method and device, electronic equipment and storage medium |
CN113434658A (en) * | 2021-08-25 | 2021-09-24 | 西安热工研究院有限公司 | Thermal power generating unit operation question-answer generation method, system, equipment and readable storage medium |
CN117573849A (en) * | 2024-01-16 | 2024-02-20 | 之江实验室 | Knowledge graph multi-hop question-answering method, device, equipment and storage medium |
CN117573849B (en) * | 2024-01-16 | 2024-04-19 | 之江实验室 | Knowledge graph multi-hop question-answering method, device, equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
HK1217789A1 (en) | 2017-01-20 |
CN105095195B (en) | 2018-09-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105095195A (en) | Method and system for human-machine questioning and answering based on knowledge graph | |
CN106776711B (en) | Chinese medical knowledge map construction method based on deep learning | |
CN107291783B (en) | Semantic matching method and intelligent equipment | |
CN112184525B (en) | System and method for realizing intelligent matching recommendation through natural semantic analysis | |
Zhu et al. | Mobile app classification with enriched contextual information | |
CN111708869B (en) | Processing method and device for man-machine conversation | |
CN111581990B (en) | Cross-border transaction matching method and device | |
CN106970991B (en) | Similar application identification method and device, application search recommendation method and server | |
WO2015027315A1 (en) | System and method for identifying and scoring leads from social media | |
US10740406B2 (en) | Matching of an input document to documents in a document collection | |
CN108305180B (en) | Friend recommendation method and device | |
CN110765761A (en) | Contract sensitive word checking method and device based on artificial intelligence and storage medium | |
CN115002200A (en) | User portrait based message pushing method, device, equipment and storage medium | |
CN112395410A (en) | Entity extraction-based industry public opinion recommendation method and device and electronic equipment | |
CN114357117A (en) | Transaction information query method and device, computer equipment and storage medium | |
CN113268615A (en) | Resource label generation method and device, electronic equipment and storage medium | |
CN114387061A (en) | Product pushing method and device, electronic equipment and readable storage medium | |
CN112507089A (en) | Intelligent question-answering engine based on knowledge graph and implementation method thereof | |
CN113627797A (en) | Image generation method and device for employee enrollment, computer equipment and storage medium | |
Leonandya et al. | A semi-supervised algorithm for Indonesian named entity recognition | |
CN109684357B (en) | Information processing method and device, storage medium and terminal | |
US20230090601A1 (en) | System and method for polarity analysis | |
US20200184521A1 (en) | Method and system for initiating an interface concurrent with generation of a transitory sentiment community | |
CN114528851B (en) | Reply sentence determination method, reply sentence determination device, electronic equipment and storage medium | |
CN113254623B (en) | Data processing method, device, server, medium and product |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 1217789 Country of ref document: HK |
|
GR01 | Patent grant | ||
GR01 | Patent grant |