CN101373532A - FAQ Chinese request-answering system implementing method in tourism field - Google Patents
FAQ Chinese request-answering system implementing method in tourism field Download PDFInfo
- Publication number
- CN101373532A CN101373532A CNA200810058660XA CN200810058660A CN101373532A CN 101373532 A CN101373532 A CN 101373532A CN A200810058660X A CNA200810058660X A CN A200810058660XA CN 200810058660 A CN200810058660 A CN 200810058660A CN 101373532 A CN101373532 A CN 101373532A
- Authority
- CN
- China
- Prior art keywords
- question
- field
- faq
- sentence
- answer
- 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
- 238000000034 method Methods 0.000 title claims abstract description 67
- 238000004458 analytical method Methods 0.000 claims abstract description 24
- 239000000284 extract Substances 0.000 claims abstract description 23
- 238000010276 construction Methods 0.000 claims abstract description 3
- 230000004927 fusion Effects 0.000 claims description 5
- 238000004891 communication Methods 0.000 claims description 4
- 230000002596 correlated effect Effects 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 abstract description 11
- 238000004364 calculation method Methods 0.000 abstract description 9
- 238000012360 testing method Methods 0.000 abstract description 8
- 230000000694 effects Effects 0.000 abstract description 5
- 230000008520 organization Effects 0.000 abstract description 2
- 230000010354 integration Effects 0.000 abstract 1
- 230000000877 morphologic effect Effects 0.000 abstract 1
- 239000000203 mixture Substances 0.000 description 9
- 230000000875 corresponding effect Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000013459 approach Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 241000531229 Caryopteris x clandonensis Species 0.000 description 2
- 235000001486 Salvia viridis Nutrition 0.000 description 2
- 150000001875 compounds Chemical class 0.000 description 2
- 238000009472 formulation Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000035800 maturation Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Abstract
The invention provides an implement method of an FAQ Chinese question answering system in the tourism field. The implement method comprises the steps of FAQ collection and organization, construction of a tourism field knowledge base, user query, question analysis, answer extraction and the like, thereby realizing the FAQ Chinese question answering system in the tourism field. The implement method constructs the tourism field knowledge base-field knowledge network with the help of the idea of ontology, utilizes the KDML language to define and describe the terms and the relations of the tourism field and realizes the integration of the tourism field knowledge network and a general knowledge base-knowledge network. The invention proposes a calculation method of similarity of tourism questions on the basis, the method realizes the calculation of the similarity of the questions with the help of the characteristics of the questions of the tourism field and the combination of morphological relations, the syntactic dependency relations and the field concept semantic relations in the questions, and the method further searches the related question from a candidate question set and extracts the answer of the question based on the similarity calculation. The test result of a Yunnan tourism FAQ question answering system proves that the method is feasible and has better effect.
Description
Technical field
The present invention relates to a kind of tour field FAQ Chinese question answering system implementation method, especially a kind of question answering system implementation method based on tour field frequently asked question storehouse (FAQ) belongs to artificial intelligence field.
Background technology
Frequently asked question FAQ (Frequently-asked Question) is the main means that online help is provided on the current network, answers questions by organizing some possible often asking in advance, is distributed on to be user's service of providing advice on the webpage.The FAQ knowledge organization is simple, easy to maintenance, but, accumulation gradually along with the frequently asked question collection, problem quantity is increasing, the knowledge acquisition approach of browsing formula page by page will more and more be difficult to satisfy user's actual demand, will waste a large amount of quality time of user, in addition when user capture just find not have the own information that really needs, time and effort consuming during all-links at all.
Summary of the invention
Purpose of the present invention provides a kind of tour field FAQ Chinese question answering system implementation method for addressing the above problem just, with efficiently, is user's service of providing advice quickly and accurately.
The present invention finishes by following technical proposal: a kind of tour field FAQ Chinese question answering system implementation method is characterized in that comprising:
(1) FAQ collects and tissue: combining artificial or semi-automatic mode, extract the tourism question and answer from the internet right, and arrangement enters tourism question and answer storehouse, forms the FAQ storehouse of travelling;
(2) tour field construction of knowledge base: make up and safeguard the tour field structure of knowledge and relation, form the tour field knowledge base;
(3) user inquiring: on the internet, the user carries out the travel information inquiry by natural language problem;
(4) case study: the problem to user input is analyzed, and it is interdependent to information such as, problem typeses to extract keyword, expansion word, the sentence structure of sign problem;
(5) answer is extracted: according to the case study result, from frequently asked question storehouse (FAQ), put forward retrieval candidate problem, adopt the field question similarity calculating method, calculate customer problem and candidate problem similarity, the problem answers that extracts the similarity maximum is as the candidate answer, and offer the user, return the final user and inquire about answer; The user can provide the natural language problem towards text, and system directly returns answer, rather than a large amount of webpages relevant with problem.
Described step (1) FAQ collects with method for organizing and is specially: climb automatically from the internet by web crawlers for first kind and get, and enter the FAQ storehouse by artificial screening; Second kind is by artificially collecting and put in order acquisition, at tour field, the special relevant issues such as relevant introduction, admission ticket, traffic such as place, sight spot, local conditions and customs, hotel of collecting, taxonomic revision with organize the FAQ question and answer to and enter the FAQ storehouse; The third then is a non-existent new question sentence by system's automatic recording user input but in the question sentence storehouse, and this class question sentence unification is saved in the question and answer history library, regularly by the manual examination and verification arrangement, the answer and the question sentence of correspondence is gone into the FAQ storehouse together.
The right storage of the question and answer of described FAQ is by setting up problem (question) and two relation tables of answer (answer), and (Questionid Answerid) carries out the answer index by major key respectively; The storage of issue table, for the ease of quick retrieval, adopt the inverted index mode to store, set up the inverted index document between speech and the question sentence, the selection of candidate question set is extracted from index file, and final result is then according to the answer answerid that deposits in the issue table directly output from database.
Described step (2) has mainly been utilized ontological thought, carries out the expansion of field term by " knowing net " of language message center Mr. Dong Zhendong of institute of the existing Chinese Academy of Sciences.
Described step (2) is specially: by ontological thought the notion of yunnan tourism field resource is accurately described, based on " knowing net ", adopt the conceptual description language KDML rule of " knowing net ", set up special yunnan tourism domain body, form resource ontology library field, yunnan tourism field and know dictyosome system, and realized the fusion that net " is known " in yunnan tourism knowledge base " net is known in the field " and general knowledge storehouse.Altogether relevant 2012 notions of tourisms such as the introduction of yunnan tourism sight spot, local conditions and customs, tourist communications are defined and describe at present, formed the yunnan tourism field and " known net ".
Because with respect to open field, be subjected to confinement to have certain domain knowledge characteristics, can reduce the difficulty of natural language processing by means of the domain knowledge relation.Ontology (Ontology) is a kind of accurate description to notion, particularly to the description of field concept, is that a kind of good domain knowledge is represented mode.Body is described the ABC architecture of delineating out a certain specific area by the standardization to notion, term and mutual relationship thereof.HowNet is a general general knowledge resource " to know net ", and it has described the notion of the word representative of Chinese and english, discloses between notion and the notion and attribute that notion had and the relation between the attribute.By ontological thought the notion of tour field resource is accurately described, based on " knowing net ", adopt the conceptual description language KDML rule of " knowing net ", set up special tour field body, form tour field resource ontology library field and know dictyosome system, and realized the fusion that net " is known " in tourism knowledge base " net is known in the field " and general knowledge storehouse.Altogether relevant 2012 notions of tourisms such as the introduction of yunnan tourism sight spot, local conditions and customs, hotel, tourist communications are defined and describe at present, form the tour field knowledge base.
Described step (3) is specially: the user can adopt the natural language mode to inquire about, and directly obtain the answer of problem by be correlated with tour field problem inquiry of internet.
Described step (4) is specially: by the question sentence analysis module problem of user's input is analyzed, mainly comprise lexical analysis, syntactic analysis and semantic analysis, lexical analysis is mainly carried out participle to problem, filter stop words, extract the inquiry core word and carry out keyword expansion by " knowing net "; Syntactic analysis is mainly to extracting the core stem of question sentence, and by Harbin Institute of Technology's parser, it is interdependent right to extract the question sentence sentence structure; The question sentence semantic information is mainly analyzed in semantic analysis, and according to the tour field characteristics, tour field problem types characteristic sentence mould rule is also extracted in definition, and realizes the identification of question sentence type by the rule match mode, specifically:
A, key to the issue speech, expansion word extract:
Key to the issue speech and expansion word are the fundamentals of sign problem, also are the bases that candidate question set retrieval and answer are extracted.By the load domains knowledge base, after to the question sentence participle, carry out a field term cutting again, realize field term cutting and part-of-speech tagging, and utilize " knowing net ", keyword is carried out the synonym expansion, form the keyword expansion speech.
B, question sentence sentence structure are interdependent to extracting:
Because complete Chinese sentence is by the trunk composition of sentence and is modified into branch and constitutes, and people often just can understand the general meaning of a sentence from the trunk composition, because there are a lot of difficulties in syntactic analysis fully at present, so when carrying out sentence similarity calculating, can embody similarity between the sentence by the right similarity of effectively arranging in pairs or groups between the sentence.So-called effectively collocation is right to being meant full sentence core word and directly existing with ... the collocation that its effective speech forms, and effectively speech is defined as verb, noun and adjective here, and it is to be determined by the part-of-speech tagging behind the participle.And the shared language technology platform LTP interface by Harbin Institute of Technology obtains sentence, and effectively collocation is right.
C, the classification of tour field question sentence
Problem types is the extract key factor of answer strategy of location answer and formulation, when question sentence is analyzed, judge whether two problems are similar, whether the problem types that at first must judge two problems is consistent, if it is consistent, just can carry out further similarity and calculate judgement, otherwise two problems can not be similar.In specific area, because professional relative fixed, therefore question sentence inquiry content-form is limited relatively, therefore can classify to problem at the question and answer business, improve the accuracy rate of similar question sentence retrieval and answer extraction with this, at tour field, extract the composition rule that has made up the variety of issue type, and by these feature identification problem typeses.
Described step (5) answer extracting method is specially: question sentence relation information such as the problem types that obtains according to the problem branch, key to the issue speech, problem expansion word, adopt lucene from frequently asked question storehouse (FAQ), to retrieve the candidate problem set, by the field question similarity calculating method, the candidate problem answers that extracts the similarity maximum is as answer, and return to the user, specifically:
The selection of A, candidate question set and question sentence index
The purpose of setting up candidate question set is to dwindle seek scope, and comparatively complicated process such as follow-up similarity calculating is all carried out in this relative small range of candidate question set, thereby improves the recall precision of system.In tourism (FAQ) question answering system, adopt the Lucene searching system to realize retrieval of candidate's problem and question sentence index.
B, tour field question sentence similarity are calculated
It is the basis that similar question sentence is searched among the FAQ that the question sentence similarity is calculated, it simultaneously also is the key that answer is extracted, it directly influences the order of accuarcy that answer is extracted, current existing multiple Chinese sentence similarity computing method are divided into Three Estate: grammer similarity, semantic similarity and pragmatic similarity usually.But these similarity calculating methods all have different defectives in the question answering system of field (FAQ).At this in conjunction with tour field question sentence characteristics, take all factors into consideration semantic distance, sentence structure dependence and the field concept semantic relation factor of speech, a kind of question sentence similarity calculating method has been proposed, this method is at first utilized the field question characteristics, carry out similar problem according to problem category and filter, reach " net is known in the field " knowledge base based on " knowing net " then, it is effectively interdependent right to adopt syntactic analysis to extract question sentence, and utilize interdependent to the notion semantic relation, realize that similarity is calculated between the tour field question sentence.The calculating committed step is as follows:
(1) semantic similarity of speech calculates
Speech is the basis of forming sentence, and sentence similarity must show by the similarity that sentence comprises between the speech, and computing method are with reference to the phrase semantic similarity calculating method of Liu Qun based on " knowing net ", and computation process is as follows
Wherein, Sim
1(C
1, C
2) be first independent adopted former similarity, the Sim of notion
2(C
1, C
2) be former similarity, the Sim of other independent justice
3(C
1, C
2) for concerning adopted former similarity and Sim
4(C
1, C
2) be the former similarity in Signifier, β i (1≤i≤4) is an adjustable parameter, and: β
1+ β
2+ β
3+ β
4=1, β
1〉=β
2〉=β
3〉=β
4, β
1〉=0.5.
(2) the question sentence sentence structure is interdependent calculates similarity
The sentence structure that can extract question sentence according to the problem syntactic analysis is interdependent right, the effective collocation that has obtained two question sentences in extraction to after, must be relatively two question sentences effectively collocation between similarity, effectively arrange in pairs or groups between question sentence to similarity in order to calculate, at first must calculate two effectively collocation between similarity relation between the equivalent, according to knowing net disambiguation annotation results, employing is based on " knowing net " Word similarity method (A step), calculate interdependent similarity respectively, get two similar mean values then and obtain two effective interdependent right similarities last two corresponding words.Calculate similarity between the question sentence according to interdependent right similarity then, for any two question sentence A and B, the question sentence sentence that A comprises is interdependent to being A
1, A
2..., A
m, the question sentence sentence that B comprises is interdependent to being B
1, B
2..., Bn, at first interdependent to being benchmark with among the question sentence A calculated interdependent to A
i(1≤i≤m) and B
j(similarity S (the A between 1≤j≤n)
i, B
j), select S (A successively according to formula (2)
i, B
j) maximum interdependent right, interdependent to be sky in the A sentence picked out an effective interdependent pair set { a of sentence
i, interdependent to being benchmark with the effective sentence of B equally, it is interdependent to B to calculate sentence
j((similarity between 1≤i≤m) is expressed as S (B for 1≤j≤n) and Ai
j, A
i), select S (B successively according to formula (3)
j, A
i) maximum interdependent right, interdependent to be sky in the B sentence picked out an effective interdependent pair set { b of sentence
j.
a
i=max(S(A
i,B
1),S(A
i,B
2),Λ,S(A
i,B
n)) (2)
b
j=max(S(B
j,A
1),S(B
j,A
2),Λ,S(B
j,A
m)) (3)
(3) tour field question sentence similarity is calculated
Judgement of question sentence type and question sentence are interdependent to after extracting carrying out, just can be according to interdependent two interdependent pair set { a that obtain that similarity is calculated
i, { b
jCarry out the calculating of question sentence similarity, computing formula is as follows:
Adopt that effective sentence structure is interdependent to be calculated carrying out the question sentence similarity, reduce the complexity of algorithm greatly, accuracy rate can obtain raising to a certain degree.This method is considered from the shallow-layer sentence structure, has considered the dependence between speech and the speech, and is more abundant to the understanding of sentence, thereby can obtain the value of sentence similarity more accurately.
The retrieval of C, similar problem and answer are extracted
After determining candidate question set, further from Candidate Set, pick out the question sentence the most similar exactly to the target question sentence.The thought of similar question sentence retrieval is the similarity between each question sentence and the target question sentence in the calculated candidate problem set, get similarity greater than the question sentence of the assign thresholds λ of system as similar question sentence, and according to the answer of this question sentence correspondence numbering (AnswerId), Automatic Extraction goes out relevant answer and returns to the user as the output result from database.According to tour field question sentence characteristics, bluebeard compound semantic information, question sentence sentence structure dependence and field concept relation are carried out similarity calculating, and concrete grammar sees before and states tour field question sentence similarity calculating method content, realize that finally the similarity of Chinese question sentence is calculated.
Described answer extracting method extracts the relevant issues of user according to field question sentence similarity calculating method, and field question sentence similarity calculating method has merged domain knowledge notion and relation (tour field term and relation), syntactic structure (sentence structure interdependent to and relation) and semantic many features such as (problem typeses) and carried out sentence similarity and calculate.
The present invention has following advantage and effect: the present invention is by means of ontological thought, based on " knowing net " general knowledge storehouse, adopt the KDML descriptive language, define and described tour field term and relation, expansion tour field term description, make up tour field knowledge base-field and know net, realized the fusion that net " is known " in " tour field is known net " and general knowledge storehouse.By morphology, sentence structure and semantic analysis user natural language problem, extract identification problem keyword, expansion word, problem category, question sentence trunk, the interdependent reciprocity question sentence sign of sentence structure, and in conjunction with domain knowledge, sentence structure dependence, semantic relation, realize the calculating of question sentence similarity, and be calculated as the basis, the relevant question sentence of retrieval from the candidate problem set with similarity, extract problem answers, the Chinese question answering system of tour field frequently asked question (FAQ) can be realized, and have efficient, quick, accurate.Yunnan tourism FAQ question answering system test result shows that this method is feasible, and effect is preferably arranged.
Description of drawings
Fig. 1 is a tourism FAQ question answering system structural drawing.
Embodiment
Embodiment
As Fig. 1, tour field FAQ Chinese question answering system implementation method provided by the invention, concrete steps are as follows:
Step 1, FAQ collects and tissue: the FAQ storehouse is the core resource of question and answer, mainly by three kinds of approach acquisitions: climb automatically from the internet by web crawlers for first kind and get, and enter the FAQ storehouse by artificial screening; Second kind is by artificially collecting and put in order acquisition, at tour field, the special relevant issues such as relevant introduction, admission ticket, traffic such as place, sight spot, local conditions and customs, hotel of collecting, taxonomic revision with organize the FAQ question and answer right; The third then is a non-existent new question sentence by system's automatic recording user input but in the question sentence storehouse, and this class question sentence unification is saved in the question and answer history library, regularly by the manual examination and verification arrangement, the answer and the question sentence of correspondence is put in storage together.
FAQ stores by database mode, for improving access speed, problem (Question) and two relation tables of answer (Answer) have been defined, wherein main storage problem of issue table (Question) and answer index information, comprise question number (QuestionId), problem (Question), problem types (QuestionType) and answer numbering (AnswerId), answer table (AnswerId) storage answer information comprises answer numbering (AnswerId) and answer (Answer).Because relation table only is used for the data storage of FAQ and extract the location of answer, retrieve fast for ease of candidate's problem, all question sentences all are index, set up speech-question sentence inverted index literary composition retaining with the speech behind the participle, the selection of candidate question set is extracted from index file, and final result is then according to the answer numbering of depositing in the issue table (AnswerId) directly location output from answer table.Because people often have new problem to add, therefore, need the new problem of often expansion in FAQ, expansion and replacement problem must judge at first whether the problem of new input has same or similar problem in FAQ, basis for estimation is to calculate the similarity of each question sentence in the target question sentence of user's input and the candidate question set, set a threshold value, if similarity is not less than specified threshold value between two question sentences, think that then this question sentence semantically is being equal to the input question sentence among the FAQ, be two kinds of sayings of same problem, do not need to expand relevant issues.If two question sentence similarities are less than assign thresholds, just represent the problem that do not have the user to ask in the existing FAQ storehouse.For this class situation, system at first records this question sentence in the question and answer history library, and regularly by the area of computer aided manual sorting, adds the new problem in the history library and corresponding answer in the FAQ storehouse and sets up increment index, thereby realize the renewal of FAQ data.
Step 2, domain knowledge base makes up: by ontological thought the notion of tour field resource is accurately described, based on " knowing net ", adopt the conceptual description language KDML rule of " knowing net ", set up special tour field body, form tour field resource ontology library field and know dictyosome system, and realized the fusion that net " is known " in tourism knowledge base " net is known in the field " and general knowledge storehouse.At present altogether tourisms such as the introduction of yunnan tourism sight spot, local conditions and customs, hotel, tourist communications 2012 notions of being correlated with are defined and describe, as: accurately being described below of notion " Shangri-la " and " The Old Town of Lijiang ":
NO.=130001
The W_C=Shangri-la
G_C=N
E_C=~be a beautiful place
W_E=xigelila
G_E=N
E_E=~is?a?beautiful?place
The DEF=place| place, the city| city, ProperName| is special, (Diqing| enlightening celebrating state), (Yunnan| Yunnan Province), (China| China)
NO.=130002
The W_C=The Old Town of Lijiang
G_C=N
E_C=~very special
W_E=Old?Town?of?Lijiang
G_E=N
E_E=~is?very?special
The DEF=place| place, ProperName| is special, city| city, past| former times, (scene| scenic spot), (lijiang| Lijing), (Yunman| Yunnan)
Step 3, user inquiring: on the internet, realize the question and answer query interface based on the Web mode, the user can be by the relevant information inquiry of travelling of natural language problem mode;
Step 4, case study: case study is that the natural language problem of user input is analyzed, it is interdependent to information such as, problem typeses, specifically to extract keyword, expansion word, the sentence structure of sign problem:
1) key to the issue speech, expansion word extract:
Key to the issue speech and expansion word are the fundamentals of sign problem, also are the bases that candidate question set retrieval and answer are extracted.Therefore, problem is carried out participle and part of speech mark, but for specific area, because field vocabulary may fail to show in general dictionary, therefore Words partition system can not well be discerned field vocabulary, cause a field vocabulary may be cut into a plurality of universal words, for this reason, by the load domains knowledge base, after to question sentence cutting just, carry out a field term cutting again, thereby well realized field term cutting and part-of-speech tagging, after carrying out word segmentation, removed stop words, extract noun, verb, adjective, limited adverbial word constitutes the key to the issue speech, and utilize " knowing net ", and keyword is carried out the synonym expansion, form the keyword expansion speech.
2) the question sentence sentence structure is interdependent to extracting:
Because complete Chinese sentence is by the trunk composition of sentence and is modified into branch and constitutes, and people often just can understand the general meaning of a sentence from the trunk composition, because there are a lot of difficulties in syntactic analysis fully at present, so when carrying out sentence similarity calculating, can embody similarity between the sentence by the right similarity of effectively arranging in pairs or groups between the sentence.So-called effectively collocation is right to being meant full sentence core word and directly existing with ... the collocation that its effective speech forms, and effectively speech is defined as verb, noun and adjective here, and it is to be determined by the part-of-speech tagging behind the participle.Full sentence core word is the root node of dependency tree.Problem Q1 for example: is there which characteristic red-letter day in the Dai nationality? with problem Q2: what red-letter day the ethnic group in Yunnan has?, wherein effectively taking of question sentence 1 is paired into: have-red-letter day, have-the Dai nationality; Effectively taking of question sentence 2 is paired into: have-red-letter day, have-ethnic group.As long as relatively these collocation between similarity degree, the interdependent shared language technology platform LTP interface to employing Harbin Institute of Technology of sentence obtains.
3) tour field question sentence classification
Problem types is the extract key factor of answer strategy of location answer and formulation, such as the problem of asking " sight spot introduction " type, just can not answer with the content of " special favor " problem types, problem types is having important status aspect similar question sentence retrieval and the answer extraction, when question sentence is analyzed, judge whether two problems are similar, whether the problem types that at first must judge two problems is consistent, if it is consistent, just can carry out further similarity and calculate judgement, otherwise two problems can not be similar.In specific area, because professional relative fixed, so question sentence inquiry content-form is limited relatively, therefore can classify to problem at the question and answer business, improve the accuracy rate of similar question sentence retrieval and answer extraction with this, at tour field, frequently asked question is divided into sight spot introduction, position, sight spot, sight spot ticket price, local delicacies, special product introduction, 23 subclass types such as custom red-letter day, and according to the feature rule of various somes problem typeses of the feature extraction of these problem typeses, such as scape
The point location type, its relevant issues form is: where is * * * * sight spot?, where * * * * sight spot is positioned at?, where * * * sight spot waits if being located in, and its type constitution rule is * * * (sight spot)+be located in/be positioned at /+interrogative (where).Extraction has made up the composition rule of variety of issue type, and by these feature identification problem typeses.
Step 5, answer is extracted: the answer extracting method mainly comprises following step:
1) selection of candidate question set and question sentence index
The purpose of setting up candidate question set is to dwindle seek scope, and comparatively complicated process such as follow-up similarity calculating is all carried out in this relative small range of candidate question set, thereby improves the recall precision of system.Because the effect of candidate question set is to concentrate from extensive question sentence to take out a fuzzy correlation fast but less relatively subclass, therefore, can the stable searching system of selective maturation realize the retrieval of candidate's problem, Lucene is as powerful, a cross-platform searching system, obtained widespread use, as the retrieval of Sogou news section, the help of Jive WEB forum, Cocoon, Eclipse part etc.The selection Lucene that increases income is used for candidate question set retrieval, and this module effectiveness of retrieval and accuracy rate can effectively be guaranteed.
Lucene is when setting up index for file, at first need to be converted into document (Document) object that to discern, each document then is made up of one or more field (Field) object, and field includes a title and corresponding value again, as a project in the hash table.In actual applications, generally all corresponding to a segment information relevant with inquiry or result for retrieval, for example, web page title need appear in the Search Results field, so can add it in the document object to as a field.Field both can be indexed, can be not indexed and directly be saved in the document yet, as for unique ID, just need not index, as long as preserve.
Because Chinese is with English different, lack dividing mark between the word, with the speech is that the Lucene that handles unit can't resolve Chinese text, therefore, following 3 crucial processing procedures have been increased, be used to realize index: at first, the input sentence is carried out word segmentation processing, thereby obtain each word in the question sentence to Chinese text.Next filters the garbage in the question sentence, mainly refers to the filtration of stop words, as filter in the sentence " ", " youngster ", " " etc., punctuation mark etc.Set up index according to the speech that obtains at last.When index building, at first create an index construct module, wherein the analyzer of the memory location of index file and index content is specified in its constructed fuction, order reads each bar record of question sentence table in the FAQ storehouse then, and be index content with the question sentence, the answer ID of question sentence ID, question sentence correspondence is that index key is set up a Lucene document object, and the document object that generates is joined in the rope structure module, and so circulation is up to all question sentences are all joined index file.
To the problem that obtains by retrieval, filter according to the target problem problem types again, from the problem set that retrieval is recalled, remove the problem that does not conform to the target problem problem types and form the candidate question set that answer is extracted.
2) tour field question sentence similarity is calculated
It is the basis that similar question sentence is searched among the FAQ that the question sentence similarity is calculated, it simultaneously also is the key that answer is extracted, it directly influences the order of accuarcy that answer is extracted, current existing multiple Chinese sentence similarity computing method are divided into Three Estate: grammer similarity, semantic similarity and pragmatic similarity usually.The pragmatic similarity has suitable difficulty, and effect is undesirable at present.And in general application, the semantic similarity that calculates sentence just can satisfy the demands substantially.At present the research method that the sentence semantic similarity is calculated mainly contains: based on the method for identical vocabulary, based on the method for semantic dictionary, based on the method for dependency tree, and based on the method for editing distance etc.Wherein, clearly limitation is arranged based on the method for identical vocabulary, then powerless for the replacement between the synonym.And the method for use semantic dictionary can well address this problem, but the method for simple use semantic dictionary is not considered the structure of sentence inside and the interaction relationship between the word, and accuracy rate is not high.Utilize between the sentence sentence structure dependence to carry out similarity based on the method for dependency tree and calculate, considered the syntax structural relationship of sentence, but faced the precision problem of complete syntactic analysis, and do not consider that the nearly justice of synonym of vocabulary in the syntactic structure replaces.Edit distance approach is normally used for the quick fuzzy matching field of sentence, but the editing operation underaction of its regulation does not consider that the synonym of word is replaced yet.
Calculate on the existing problems basis in the parsing sentence similarity, in conjunction with tour field question sentence characteristics, take all factors into consideration semantic distance, sentence structure dependence and the field concept semantic relation factor of speech, a kind of question sentence similarity calculating method has been proposed, this method is at first utilized the field question characteristics, carrying out similar problem according to problem category filters, reach " net is known in the field " knowledge base based on " knowing net " then, it is effectively interdependent right to adopt syntactic analysis to extract question sentence, and utilize interdependent to the notion semantic relation, realize that similarity is calculated between the tour field question sentence.The calculating committed step is as follows:
A. the semantic similarity of speech calculates
Speech is the basis of forming sentence, sentence similarity must show by the similarity that sentence comprises between the speech, in actual applications, it is just the same to tend to occur two question sentence meanings, but its expression-form is different, such as, what sight spot does problem Q3: the Shangri-la have? with problem Q4: there are those joyful places in middle pasture? its main cause is because the synonym and the correlationship of speech cause, the speech that occurs in a question sentence exists certain synonym and related term, therefore, when calculating the word similarity, must consider the synonym and the correlationship of speech, and can not only judge according to the superficial feature of speech own, " net is known in the field " that utilizes " knowing net " and expand in the field carries out the calculation of similarity degree of the word-level of question sentence, by question sentence being known the net disambiguation, and utilize the notion that occurs in the question sentence to calculate similarity between the question sentence, semantic distance between the notion is defined as the adopted former bee-line in adopted elite tree of two notion correspondences, computing method are with reference to the phrase semantic similarity calculating method of Liu Qun based on " knowing net ", and computation process is as follows
Wherein, Sim
1(C
1, C
2) be first independent adopted former similarity, the Sim of notion
2(C
1, C
2) be former similarity, the Sim of other independent justice
3(C
1, C
2) for concerning adopted former similarity and Sim
4(C
1, C
2) be the former similarity in Signifier, β i (1≤i≤4) is an adjustable parameter, and: β
1+ β
2+ β
3+ β
4=1, β
1〉=β
2〉=β
3〉=β
4, β
1〉=0.5.
B. the question sentence sentence structure is interdependent calculates similarity
The sentence structure that can extract question sentence according to the problem syntactic analysis is interdependent right, the effective collocation that has obtained two question sentences in extraction to after, must be relatively two question sentences effectively collocation between similarity, effectively arrange in pairs or groups between question sentence to similarity in order to calculate, at first must calculate two effectively collocation between similarity relation between the equivalent, according to knowing net disambiguation annotation results, employing is based on " knowing net " Word similarity method (A step), calculate interdependent similarity respectively to last two corresponding words, calculate verb respectively such as interdependent similarity similarity with " having " " is arranged " " having-the Dai nationality " and " having-ethnic group ", " the Dai nationality " and " ethnic group " between similarity, get two similar mean values then and obtain two effective interdependent right similarities.Calculate similarity between the question sentence according to interdependent right similarity then, for any two question sentence A and B, the question sentence sentence that A comprises is interdependent to being A
1, A
2..., A
m, the question sentence sentence that B comprises is interdependent to being B
1, B
2..., Bn, at first interdependent to being benchmark with among the question sentence A calculated interdependent to A
i(1≤i≤m) and B
j(similarity S (the A between 1≤j≤n)
i, B
j), select S (A successively according to formula (2)
i, B
j) maximum interdependent right, interdependent to be sky in the A sentence picked out an effective interdependent pair set { a of sentence
i, interdependent to being benchmark with the effective sentence of B equally, it is interdependent to B to calculate sentence
j((similarity between 1≤i≤m) is expressed as S (B for 1≤j≤n) and Ai
j, A
i), select S (B successively according to formula (3)
j, A
i) maximum interdependent right, interdependent to be sky in the B sentence picked out an effective interdependent pair set { b of sentence
j.
a
i=max(S(A
i,B
1),S(A
i,B
2),Λ,S(A
i,B
n)) (2)
b
j=max(S(B
j,A
1),S(B
j,A
2),Λ,S(B
j,A
m)) (3)
C. tour field question sentence similarity is calculated
Judgement of question sentence type and question sentence are interdependent to after extracting carrying out, just can be according to interdependent two interdependent pair set { a that obtain that similarity is calculated
i, { b
jCarry out the calculating of question sentence similarity, computing formula is as follows:
Adopt that effective sentence structure is interdependent to be calculated carrying out the question sentence similarity, reduce the complexity of algorithm greatly, accuracy rate can obtain raising to a certain degree.This method is considered from the shallow-layer sentence structure, has considered the dependence between speech and the speech, and is more abundant to the understanding of sentence, thereby can obtain the value of sentence similarity more accurately.But existing syntactic analysis technology is ripe not enough, also all syntactic information features all can't be taken into account, so calculating can produce certain error.
3) retrieval of similar problem and answer are extracted
After determining candidate question set, further from Candidate Set, pick out the question sentence the most similar exactly to the target question sentence.The thought of similar question sentence retrieval is the similarity between each question sentence and the target question sentence in the calculated candidate problem set, get similarity greater than the question sentence of the assign thresholds λ of system as similar question sentence, and according to the answer of this question sentence correspondence numbering (AnswerId), Automatic Extraction goes out relevant answer and returns to the user as the output result from database.According to tour field question sentence characteristics, bluebeard compound semantic information, question sentence sentence structure dependence and field concept relation are carried out similarity calculating, and concrete grammar sees before and states tour field question sentence similarity calculating method content, realize that finally the similarity of Chinese question sentence is calculated.
Consider the calculation of similarity degree error, system can also provide preceding 4 suboptimum records when providing optimum answer, selects for the user, if optimum answer is really not relevant, the user can also further search from these candidate answers.From user's angle, by relevant issues being determined to fast among the less set, the practical value of system has just embodied.
The experiment of yunnan tourism FAQ question answering system
At the yunnan tourism field, made up the tour field knowledge base, expanded 2012 field concepts by " knowing net " and obtained " net is known in the field ", adopt retrieval mode with artificial and the Automatic Extraction mode is collected and organized 23335 question and answer right, extract 188 different question sentence type feature rules, adopted the Web mode to realize yunnan tourism FAQ question answering system.At present, this system has been carried out the test of two aspects: be question and answer test on the one hand towards the question sentence corpus, it is right wherein to have collected altogether in the question sentence corpus about 23335 question and answer of yunnan tourism, stores in FAQ, chooses 600 question sentences arbitrarily and test from the question sentence corpus; Be towards actual user's on-the-spot test on the other hand, organized 10 visitor's random enquire problems.Experimental result is as shown in table 1.
Table 1: yunnan tourism FAQ question and answer prototype system test result
Category of test | The problem number/ | Correctly reply/ | Wrong responses/ | Do not reply/ | Accurate rate % | Recall rate % |
1 | 600 | 555 | 24 | 21 | 92.5 | 96.8 |
2 | 300 | 250 | 13 | 47 | 83.3 | 87.5 |
From the actual experiment result, from the actual experiment result, She Ji yunnan tourism FAQ question answering system can be practical by this method by this.
Claims (10)
1. tour field FAQ Chinese question answering system implementation method is characterized in that comprising:
(1) FAQ collects and tissue: combining artificial or semi-automatic mode, extract the tourism question and answer from the internet right, and arrangement enters tourism question and answer storehouse, forms the FAQ storehouse of travelling;
(2) tour field construction of knowledge base: make up and safeguard the tour field structure of knowledge and relation, form the tour field knowledge base;
(3) user inquiring: on the internet, the user carries out the travel information inquiry by natural language problem;
(4) case study: the problem to user input is analyzed, and it is interdependent to information such as, problem typeses to extract keyword, expansion word, the sentence structure of sign problem;
(5) answer is extracted: according to the case study result, from the FAQ of frequently asked question storehouse, put forward retrieval candidate problem, adopt the field question similarity calculating method, calculate customer problem and candidate problem similarity, the problem answers that extracts the similarity maximum is as the candidate answer, and offer the user, return the final user and inquire about answer.
2. tour field FAQ Chinese question answering system implementation method according to claim 1 is characterized in that the user can provide the natural language problem towards text, and system directly returns answer, rather than a large amount of webpages relevant with problem.
3. tour field FAQ Chinese question answering system implementation method according to claim 1, it is characterized in that, described step (1) FAQ collects with method for organizing and is specially: climb automatically from the internet by web crawlers for first kind and get, and enter the FAQ storehouse by artificial screening; Second kind is by artificially collecting and put in order acquisition, at tour field, the special relevant issues such as relevant introduction, admission ticket, traffic such as place, sight spot, local conditions and customs, hotel of collecting, taxonomic revision with organize the FAQ question and answer to and enter the FAQ storehouse; The third then is a non-existent new question sentence by system's automatic recording user input but in the question sentence storehouse, and this class question sentence unification is saved in the question and answer history library, regularly by the manual examination and verification arrangement, the answer and the question sentence of correspondence is gone into the FAQ storehouse together.
4. tour field FAQ Chinese question answering system implementation method according to claim 3, it is characterized in that, the right storage of the question and answer of FAQ is by setting up two relation tables of problem question and answer answer, and by major key Questionid, Answerid carries out the answer index respectively; The storage of issue table, for the ease of quick retrieval, adopt the inverted index mode to store, set up the inverted index document between speech and the question sentence, the selection of candidate question set is extracted from index file, and final result is then according to the answer answerid that deposits in the issue table directly output from database.
5. tour field FAQ Chinese question answering system implementation method according to claim 1, it is characterized in that, described step (2) has mainly been utilized ontological thought, carries out the expansion of field term by " knowing net " of language message center Mr. Dong Zhendong of institute of the existing Chinese Academy of Sciences.
6. tour field FAQ Chinese question answering system implementation method according to claim 1, it is characterized in that, described step (2) is specially: by ontological thought the notion of yunnan tourism field resource is accurately described, based on " knowing net ", adopt the conceptual description language KDML rule of " knowing net ", set up special yunnan tourism domain body, form resource ontology library field, yunnan tourism field and know dictyosome system, and realized the fusion that net " is known " in yunnan tourism knowledge base " net is known in the field " and general knowledge storehouse.Altogether relevant 2012 notions of tourisms such as the introduction of yunnan tourism sight spot, local conditions and customs, tourist communications are defined and describe at present, formed the yunnan tourism field and " known net ".
7. tour field FAQ Chinese question answering system implementation method according to claim 1, it is characterized in that, described step (3) is specially: user inquiring provides user and natural language mode to put question to, and the user can be by be correlated with tour field problem inquiry of internet.
8. tour field FAQ Chinese question answering system implementation method according to claim 1, it is characterized in that, described step (4) is specially: by the question sentence analysis module problem of user's input is analyzed, mainly comprise lexical analysis, syntactic analysis and semantic analysis, lexical analysis is mainly carried out participle to problem, filter stop words, extract the inquiry core word and carry out keyword expansion by " knowing net "; Syntactic analysis is mainly to extracting the core stem of question sentence, and by Harbin Institute of Technology's parser, it is interdependent right to extract the question sentence sentence structure; The question sentence semantic information is mainly analyzed in semantic analysis, and according to the tour field characteristics, definition is also extracted tour field problem types characteristic sentence mould rule, and realizes the identification of question sentence type by the rule match mode.
9. tour field FAQ Chinese question answering system implementation method according to claim 1, it is characterized in that, described step (5) answer extracting method is specially: question sentence relation information such as the problem types that obtains according to the problem branch, key to the issue speech, problem expansion word, adopt lucene from the FAQ of frequently asked question storehouse, to retrieve the candidate problem set, by the field question similarity calculating method, the candidate problem answers that extracts the similarity maximum is as answer, and returns to the user.
10. tour field FAQ Chinese question answering system implementation method according to claim 9, it is characterized in that the answer extracting method extracts the relevant issues of user according to field question sentence similarity calculating method, field question sentence similarity calculating method merged domain knowledge notion and relation be tour field term and relation, syntactic structure be sentence structure interdependent to and relation and semanteme be that many features such as problem types are carried out sentence similarity and calculated.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNA200810058660XA CN101373532A (en) | 2008-07-10 | 2008-07-10 | FAQ Chinese request-answering system implementing method in tourism field |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNA200810058660XA CN101373532A (en) | 2008-07-10 | 2008-07-10 | FAQ Chinese request-answering system implementing method in tourism field |
Publications (1)
Publication Number | Publication Date |
---|---|
CN101373532A true CN101373532A (en) | 2009-02-25 |
Family
ID=40447685
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CNA200810058660XA Pending CN101373532A (en) | 2008-07-10 | 2008-07-10 | FAQ Chinese request-answering system implementing method in tourism field |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101373532A (en) |
Cited By (101)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102012936A (en) * | 2010-12-07 | 2011-04-13 | 中国电信股份有限公司 | Massive data aggregation method and system based on cloud computing platform |
CN102637192A (en) * | 2012-02-17 | 2012-08-15 | 清华大学 | Method for answering with natural language |
CN102662929A (en) * | 2012-03-20 | 2012-09-12 | 中南大学 | Method and device for intelligent problem analyzing and processing based on ontology |
CN102681992A (en) * | 2011-03-07 | 2012-09-19 | 腾讯科技(深圳)有限公司 | Method and system for data hierarchy |
CN102929891A (en) * | 2011-08-11 | 2013-02-13 | 阿里巴巴集团控股有限公司 | Text processing method and device |
CN103294725A (en) * | 2012-03-03 | 2013-09-11 | 李辉 | Intelligent response robot software |
CN103440287A (en) * | 2013-08-14 | 2013-12-11 | 广东工业大学 | Web question-answering retrieval system based on product information structuring |
CN103778122A (en) * | 2012-10-17 | 2014-05-07 | 腾讯科技(深圳)有限公司 | Searching method and system |
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 |
CN104050256A (en) * | 2014-06-13 | 2014-09-17 | 西安蒜泥电子科技有限责任公司 | Initiative study-based questioning and answering method and questioning and answering system adopting initiative study-based questioning and answering method |
CN104391969A (en) * | 2014-12-04 | 2015-03-04 | 百度在线网络技术(北京)有限公司 | User query statement syntactic structure determining method and device |
CN104424290A (en) * | 2013-09-02 | 2015-03-18 | 佳能株式会社 | Voice based question-answering system and method for interactive voice system |
CN104461525A (en) * | 2014-11-27 | 2015-03-25 | 韩慧健 | Intelligent user-defined consulting platform generating system |
CN104504119A (en) * | 2014-12-30 | 2015-04-08 | 天津迈沃医药技术有限公司 | Self-analysis method of associated medicines based on medical ontology database |
CN104573028A (en) * | 2015-01-14 | 2015-04-29 | 百度在线网络技术(北京)有限公司 | Intelligent question-answer implementing method and system |
CN104834651A (en) * | 2014-02-12 | 2015-08-12 | 北京京东尚科信息技术有限公司 | Method and apparatus for providing answers to frequently asked questions |
CN104978396A (en) * | 2015-06-02 | 2015-10-14 | 百度在线网络技术(北京)有限公司 | Knowledge database based question and answer generating method and apparatus |
CN105224519A (en) * | 2014-06-27 | 2016-01-06 | 英特尔公司 | Automatic question sentence in natural language detects |
CN105224683A (en) * | 2015-10-28 | 2016-01-06 | 北京护航科技有限公司 | A kind of natural language analysis intelligent interactive method and device |
CN105302859A (en) * | 2015-09-21 | 2016-02-03 | 上海智臻智能网络科技股份有限公司 | Intelligent interaction system based on Internet |
CN105488039A (en) * | 2014-09-15 | 2016-04-13 | 华为技术有限公司 | Query method and device |
CN105550361A (en) * | 2015-12-31 | 2016-05-04 | 上海智臻智能网络科技股份有限公司 | Log processing method and apparatus, and ask-answer information processing method and apparatus |
CN105653671A (en) * | 2015-12-29 | 2016-06-08 | 畅捷通信息技术股份有限公司 | Similar information recommendation method and system |
CN105701253A (en) * | 2016-03-04 | 2016-06-22 | 南京大学 | Chinese natural language interrogative sentence semantization knowledge base automatic question-answering method |
CN105760359A (en) * | 2014-11-21 | 2016-07-13 | 财团法人工业技术研究院 | Question processing system and method thereof |
CN105824933A (en) * | 2016-03-18 | 2016-08-03 | 苏州大学 | Automatic question-answering system based on theme-rheme positions and realization method of automatic question answering system |
CN105843897A (en) * | 2016-03-23 | 2016-08-10 | 青岛海尔软件有限公司 | Vertical domain-oriented intelligent question and answer system |
CN105893560A (en) * | 2016-03-31 | 2016-08-24 | 乐视控股(北京)有限公司 | Method and device for feeding effective information back to user |
CN106095956A (en) * | 2016-06-15 | 2016-11-09 | 北京智能管家科技有限公司 | support information fission querying method and device |
CN106227788A (en) * | 2016-07-20 | 2016-12-14 | 浪潮软件集团有限公司 | Database query method based on Lucene |
CN106326422A (en) * | 2016-08-24 | 2017-01-11 | 北京大学 | Method and system for retrieving food security data information based on knowledge ontology |
CN106339366A (en) * | 2016-08-08 | 2017-01-18 | 北京百度网讯科技有限公司 | Method and device for requirement identification based on artificial intelligence (AI) |
CN106547785A (en) * | 2015-09-22 | 2017-03-29 | 阿里巴巴集团控股有限公司 | Information getting method and system in knowledge base |
CN106610934A (en) * | 2016-07-08 | 2017-05-03 | 四川用联信息技术有限公司 | Novel semantic similarity solving method in intelligent manufacturing industry |
CN106649394A (en) * | 2015-11-03 | 2017-05-10 | 中兴通讯股份有限公司 | Fusion knowledge base processing method and device and knowledge base management system |
CN106653016A (en) * | 2016-10-28 | 2017-05-10 | 上海智臻智能网络科技股份有限公司 | Intelligent interaction method and intelligent interaction device |
CN106776797A (en) * | 2016-11-22 | 2017-05-31 | 中国人名解放军理工大学 | A kind of knowledge Q-A system and its method of work based on ontology inference |
CN106777274A (en) * | 2016-06-16 | 2017-05-31 | 北京理工大学 | A kind of Chinese tour field knowledge mapping construction method and system |
CN106844741A (en) * | 2017-02-13 | 2017-06-13 | 哈尔滨工业大学 | A kind of answer method towards specific area |
CN106909930A (en) * | 2015-12-23 | 2017-06-30 | 神州数码信息系统有限公司 | A kind of model and method of the man-machine automatic switchover based on government affairs machine question answering system |
CN106951558A (en) * | 2017-03-31 | 2017-07-14 | 广东睿盟计算机科技有限公司 | A kind of data processing method of the tax intelligent consulting platform based on deep search |
CN106980652A (en) * | 2017-03-03 | 2017-07-25 | 竹间智能科技(上海)有限公司 | Intelligent answer method and system |
CN106997376A (en) * | 2017-02-28 | 2017-08-01 | 浙江大学 | The problem of one kind is based on multi-stage characteristics and answer sentence similarity calculating method |
CN107103005A (en) * | 2016-02-23 | 2017-08-29 | 阿里巴巴集团控股有限公司 | The collection method and device of question and answer language material |
CN107122421A (en) * | 2017-04-05 | 2017-09-01 | 北京大学 | Information retrieval method and device |
CN107193872A (en) * | 2017-04-14 | 2017-09-22 | 深圳前海微众银行股份有限公司 | Question and answer data processing method and device |
CN107256226A (en) * | 2017-04-28 | 2017-10-17 | 北京神州泰岳软件股份有限公司 | The construction method and device of a kind of knowledge base |
CN107256227A (en) * | 2017-04-28 | 2017-10-17 | 北京神州泰岳软件股份有限公司 | Towards the semantic concept spread generating method and device of knowledge content |
CN107329967A (en) * | 2017-05-12 | 2017-11-07 | 北京邮电大学 | Question answering system and method based on deep learning |
CN107391614A (en) * | 2017-07-04 | 2017-11-24 | 重庆智慧思特大数据有限公司 | A kind of Chinese question and answer matching process based on WMD |
CN107391706A (en) * | 2017-07-28 | 2017-11-24 | 湖北文理学院 | A kind of city tour's question answering system based on mobile Internet |
CN107436916A (en) * | 2017-06-15 | 2017-12-05 | 百度在线网络技术(北京)有限公司 | The method and device of intelligent prompt answer |
WO2018000277A1 (en) * | 2016-06-29 | 2018-01-04 | 深圳狗尾草智能科技有限公司 | Question and answer method and system, and robot |
CN107562789A (en) * | 2017-07-28 | 2018-01-09 | 深圳前海微众银行股份有限公司 | Knowledge base problem update method, customer service robot and readable storage medium storing program for executing |
CN107679231A (en) * | 2017-10-24 | 2018-02-09 | 济南浪潮高新科技投资发展有限公司 | A kind of vertical field and the implementation method of Opening field mixed type intelligent Answer System |
CN107992539A (en) * | 2017-11-24 | 2018-05-04 | 合肥博焱智能科技有限公司 | Scenic spot tour-guide inquiring system based on artificial intelligence |
CN107992518A (en) * | 2017-10-31 | 2018-05-04 | 厦门快商通信息技术有限公司 | A kind of automatic question-answering method and system for lifting Intelligent dialogue speed |
CN108280218A (en) * | 2018-02-07 | 2018-07-13 | 逸途(北京)科技有限公司 | A kind of flow system based on retrieval and production mixing question and answer |
CN108334490A (en) * | 2017-04-07 | 2018-07-27 | 腾讯科技(深圳)有限公司 | Keyword extracting method and keyword extracting device |
CN108345640A (en) * | 2018-01-12 | 2018-07-31 | 上海大学 | A kind of question and answer building of corpus method based on neural network semantic analysis |
CN108416060A (en) * | 2018-03-23 | 2018-08-17 | 芜湖乐锐思信息咨询有限公司 | A kind of city tour's question answering system based on mobile interchange |
CN105592234B (en) * | 2014-10-21 | 2018-08-28 | 阿里巴巴集团控股有限公司 | Processing method of communication data and device |
CN108509617A (en) * | 2018-04-04 | 2018-09-07 | 上海智臻智能网络科技股份有限公司 | Construction of knowledge base, intelligent answer method and device, storage medium, the terminal in knowledge based library |
WO2018157805A1 (en) * | 2017-03-03 | 2018-09-07 | 腾讯科技(深圳)有限公司 | Automatic questioning and answering processing method and automatic questioning and answering system |
CN108846292A (en) * | 2018-05-30 | 2018-11-20 | 中国联合网络通信集团有限公司 | Desensitization process method and device |
CN108984778A (en) * | 2018-07-25 | 2018-12-11 | 南京瓦尔基里网络科技有限公司 | A kind of intelligent interaction automatically request-answering system and self-teaching method |
CN109101579A (en) * | 2018-07-19 | 2018-12-28 | 深圳追科技有限公司 | customer service robot knowledge base ambiguity detection method |
CN109241266A (en) * | 2015-07-23 | 2019-01-18 | 上海智臻智能网络科技股份有限公司 | The creation method and apparatus that extension is asked are asked based on standard in human-computer interaction |
CN109271503A (en) * | 2018-11-06 | 2019-01-25 | 北京猎户星空科技有限公司 | Intelligent answer method, apparatus, equipment and storage medium |
CN109344385A (en) * | 2018-01-30 | 2019-02-15 | 深圳壹账通智能科技有限公司 | Natural language processing method, apparatus, computer equipment and storage medium |
CN109359302A (en) * | 2018-10-26 | 2019-02-19 | 重庆大学 | A kind of optimization method of field term vector and fusion sort method based on it |
CN109446304A (en) * | 2018-10-10 | 2019-03-08 | 长沙师范学院 | Intelligent customer service session method and system |
CN109446305A (en) * | 2018-10-10 | 2019-03-08 | 长沙师范学院 | The construction method and system of intelligent tour customer service system |
CN109460448A (en) * | 2018-08-31 | 2019-03-12 | 厦门快商通信息技术有限公司 | It is a kind of can autonomous configuration FAQ service framework |
CN109460502A (en) * | 2018-09-14 | 2019-03-12 | 广州神马移动信息科技有限公司 | Answer clustering method and its device, electronic equipment, computer-readable medium |
CN109460452A (en) * | 2018-10-10 | 2019-03-12 | 长沙师范学院 | Intelligent customer service system based on ontology |
CN109582761A (en) * | 2018-09-21 | 2019-04-05 | 浙江师范大学 | A kind of Chinese intelligent Answer System method of the Words similarity based on the network platform |
CN109614464A (en) * | 2018-10-31 | 2019-04-12 | 阿里巴巴集团控股有限公司 | Method and device for traffic issues identification |
CN109710746A (en) * | 2018-12-28 | 2019-05-03 | 北京猎户星空科技有限公司 | Information processing method, device and electronic equipment |
CN110188174A (en) * | 2019-04-19 | 2019-08-30 | 浙江工业大学 | A kind of professional domain FAQ intelligent answer method excavated based on specialized vocabulary |
CN110263051A (en) * | 2019-06-11 | 2019-09-20 | 出门问问信息科技有限公司 | Question and answer for question answering system are to update method, device, equipment and storage medium |
CN110275937A (en) * | 2019-05-14 | 2019-09-24 | 闽江学院 | A kind of construction method and device of question and answer robot |
CN110309509A (en) * | 2019-06-28 | 2019-10-08 | 神思电子技术股份有限公司 | A kind of semantic knowledge base construction method |
CN110362651A (en) * | 2019-06-11 | 2019-10-22 | 华南师范大学 | Dialogue method, system, device and the storage medium that retrieval and generation combine |
CN110377700A (en) * | 2019-07-01 | 2019-10-25 | 厦门美域中央信息科技有限公司 | A kind of professional knowledge semantic retrieval system |
CN110737755A (en) * | 2018-07-03 | 2020-01-31 | 百度在线网络技术(北京)有限公司 | Retrieval method and device |
CN110750543A (en) * | 2018-07-05 | 2020-02-04 | 优视科技(中国)有限公司 | Man-machine interaction method, device, system, storage medium and electronic equipment |
CN110851576A (en) * | 2019-10-16 | 2020-02-28 | 迈达斯智能(深圳)有限公司 | Question and answer processing method, device, equipment and readable medium |
CN111198940A (en) * | 2019-12-27 | 2020-05-26 | 北京百度网讯科技有限公司 | FAQ method, question-answer search system, electronic device, and storage medium |
CN111506721A (en) * | 2020-04-22 | 2020-08-07 | 福州大学 | Question-answering system and construction method for domain knowledge graph |
CN111813911A (en) * | 2020-06-30 | 2020-10-23 | 神思电子技术股份有限公司 | Knowledge automatic acquisition and updating system based on user supervision feedback and working method thereof |
CN111930911A (en) * | 2020-08-12 | 2020-11-13 | 杭州东方通信软件技术有限公司 | Rapid field question-answering method and device |
CN111984774A (en) * | 2020-08-11 | 2020-11-24 | 北京百度网讯科技有限公司 | Search method, device, equipment and storage medium |
CN112527965A (en) * | 2020-12-18 | 2021-03-19 | 国家电网有限公司客户服务中心 | Automatic question answering implementation method and device based on combination of professional library and chatting library |
CN112749265A (en) * | 2021-01-08 | 2021-05-04 | 哈尔滨工业大学 | Intelligent question-answering system based on multiple information sources |
CN112925915A (en) * | 2019-12-06 | 2021-06-08 | 株式会社日立制作所 | Data generation method and device in database |
CN113239161A (en) * | 2021-05-08 | 2021-08-10 | 广州华多网络科技有限公司 | Multilingual customer service response method and corresponding device, equipment and medium thereof |
CN113360626A (en) * | 2021-07-02 | 2021-09-07 | 北京容联七陌科技有限公司 | Multi-scene mixed question-answer recommendation method for intelligent customer service robot |
CN114757208A (en) * | 2022-06-10 | 2022-07-15 | 荣耀终端有限公司 | Question and answer matching method and device |
CN116578724A (en) * | 2023-07-14 | 2023-08-11 | 杭州朗目达信息科技有限公司 | Knowledge base knowledge structure construction method and device, storage medium and terminal |
CN117132392A (en) * | 2023-10-23 | 2023-11-28 | 蓝色火焰科技成都有限公司 | Vehicle loan fraud risk early warning method and system |
-
2008
- 2008-07-10 CN CNA200810058660XA patent/CN101373532A/en active Pending
Cited By (144)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102012936A (en) * | 2010-12-07 | 2011-04-13 | 中国电信股份有限公司 | Massive data aggregation method and system based on cloud computing platform |
CN102012936B (en) * | 2010-12-07 | 2013-04-17 | 中国电信股份有限公司 | Massive data aggregation method and system based on cloud computing platform |
CN102681992A (en) * | 2011-03-07 | 2012-09-19 | 腾讯科技(深圳)有限公司 | Method and system for data hierarchy |
CN102929891B (en) * | 2011-08-11 | 2015-09-16 | 阿里巴巴集团控股有限公司 | The method and apparatus of process text |
CN102929891A (en) * | 2011-08-11 | 2013-02-13 | 阿里巴巴集团控股有限公司 | Text processing method and device |
CN102637192A (en) * | 2012-02-17 | 2012-08-15 | 清华大学 | Method for answering with natural language |
CN103294725A (en) * | 2012-03-03 | 2013-09-11 | 李辉 | Intelligent response robot software |
CN102662929A (en) * | 2012-03-20 | 2012-09-12 | 中南大学 | Method and device for intelligent problem analyzing and processing based on ontology |
CN103778122A (en) * | 2012-10-17 | 2014-05-07 | 腾讯科技(深圳)有限公司 | Searching method and system |
CN103778122B (en) * | 2012-10-17 | 2018-01-23 | 腾讯科技(深圳)有限公司 | Searching method and system |
CN103440287A (en) * | 2013-08-14 | 2013-12-11 | 广东工业大学 | Web question-answering retrieval system based on product information structuring |
CN103440287B (en) * | 2013-08-14 | 2016-12-28 | 广东工业大学 | A kind of Web question and answer searching system based on product information structure |
CN104424290A (en) * | 2013-09-02 | 2015-03-18 | 佳能株式会社 | Voice based question-answering system and method for interactive voice system |
CN104834651A (en) * | 2014-02-12 | 2015-08-12 | 北京京东尚科信息技术有限公司 | Method and apparatus for providing answers to frequently asked questions |
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 |
CN104050256A (en) * | 2014-06-13 | 2014-09-17 | 西安蒜泥电子科技有限责任公司 | Initiative study-based questioning and answering method and questioning and answering system adopting initiative study-based questioning and answering method |
CN104050256B (en) * | 2014-06-13 | 2017-05-24 | 西安蒜泥电子科技有限责任公司 | Initiative study-based questioning and answering method and questioning and answering system adopting initiative study-based questioning and answering method |
CN105224519B (en) * | 2014-06-27 | 2019-08-16 | 英特尔公司 | Automatic question sentence detection in natural language |
CN105224519A (en) * | 2014-06-27 | 2016-01-06 | 英特尔公司 | Automatic question sentence in natural language detects |
CN105488039A (en) * | 2014-09-15 | 2016-04-13 | 华为技术有限公司 | Query method and device |
CN105592234B (en) * | 2014-10-21 | 2018-08-28 | 阿里巴巴集团控股有限公司 | Processing method of communication data and device |
CN105760359B (en) * | 2014-11-21 | 2020-03-20 | 财团法人工业技术研究院 | Question processing system and method thereof |
CN105760359A (en) * | 2014-11-21 | 2016-07-13 | 财团法人工业技术研究院 | Question processing system and method thereof |
CN104461525B (en) * | 2014-11-27 | 2018-01-23 | 韩慧健 | A kind of intelligent consulting platform generation system that can customize |
CN104461525A (en) * | 2014-11-27 | 2015-03-25 | 韩慧健 | Intelligent user-defined consulting platform generating system |
CN104391969A (en) * | 2014-12-04 | 2015-03-04 | 百度在线网络技术(北京)有限公司 | User query statement syntactic structure determining method and device |
CN104391969B (en) * | 2014-12-04 | 2018-01-30 | 百度在线网络技术(北京)有限公司 | Determine the method and device of user's query statement syntactic structure |
CN104504119B (en) * | 2014-12-30 | 2018-02-02 | 天津迈沃医药技术股份有限公司 | The autoanalysis method of association medicine based on medical information ontology database |
CN104504119A (en) * | 2014-12-30 | 2015-04-08 | 天津迈沃医药技术有限公司 | Self-analysis method of associated medicines based on medical ontology database |
CN104573028A (en) * | 2015-01-14 | 2015-04-29 | 百度在线网络技术(北京)有限公司 | Intelligent question-answer implementing method and system |
US10242049B2 (en) | 2015-01-14 | 2019-03-26 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method, system and storage medium for implementing intelligent question answering |
CN104978396A (en) * | 2015-06-02 | 2015-10-14 | 百度在线网络技术(北京)有限公司 | Knowledge database based question and answer generating method and apparatus |
CN109241266A (en) * | 2015-07-23 | 2019-01-18 | 上海智臻智能网络科技股份有限公司 | The creation method and apparatus that extension is asked are asked based on standard in human-computer interaction |
CN105302859A (en) * | 2015-09-21 | 2016-02-03 | 上海智臻智能网络科技股份有限公司 | Intelligent interaction system based on Internet |
CN105302859B (en) * | 2015-09-21 | 2018-11-30 | 上海智臻智能网络科技股份有限公司 | A kind of intelligent interactive system Internet-based |
CN106547785A (en) * | 2015-09-22 | 2017-03-29 | 阿里巴巴集团控股有限公司 | Information getting method and system in knowledge base |
CN106547785B (en) * | 2015-09-22 | 2020-08-04 | 阿里巴巴集团控股有限公司 | Method and system for acquiring information in knowledge base |
CN105224683A (en) * | 2015-10-28 | 2016-01-06 | 北京护航科技有限公司 | A kind of natural language analysis intelligent interactive method and device |
CN106649394A (en) * | 2015-11-03 | 2017-05-10 | 中兴通讯股份有限公司 | Fusion knowledge base processing method and device and knowledge base management system |
WO2017076263A1 (en) * | 2015-11-03 | 2017-05-11 | 中兴通讯股份有限公司 | Method and device for integrating knowledge bases, knowledge base management system and storage medium |
CN106909930A (en) * | 2015-12-23 | 2017-06-30 | 神州数码信息系统有限公司 | A kind of model and method of the man-machine automatic switchover based on government affairs machine question answering system |
CN105653671A (en) * | 2015-12-29 | 2016-06-08 | 畅捷通信息技术股份有限公司 | Similar information recommendation method and system |
CN105550361B (en) * | 2015-12-31 | 2018-11-09 | 上海智臻智能网络科技股份有限公司 | Log processing method and device and question and answer information processing method and device |
CN105550361A (en) * | 2015-12-31 | 2016-05-04 | 上海智臻智能网络科技股份有限公司 | Log processing method and apparatus, and ask-answer information processing method and apparatus |
CN107103005B (en) * | 2016-02-23 | 2020-10-30 | 创新先进技术有限公司 | Question and answer corpus collection method and device |
CN107103005A (en) * | 2016-02-23 | 2017-08-29 | 阿里巴巴集团控股有限公司 | The collection method and device of question and answer language material |
CN105701253A (en) * | 2016-03-04 | 2016-06-22 | 南京大学 | Chinese natural language interrogative sentence semantization knowledge base automatic question-answering method |
CN105701253B (en) * | 2016-03-04 | 2019-03-26 | 南京大学 | The knowledge base automatic question-answering method of Chinese natural language question semanteme |
CN105824933B (en) * | 2016-03-18 | 2019-02-26 | 苏州大学 | Automatically request-answering system and its implementation based on main rheme |
CN105824933A (en) * | 2016-03-18 | 2016-08-03 | 苏州大学 | Automatic question-answering system based on theme-rheme positions and realization method of automatic question answering system |
CN105843897B (en) * | 2016-03-23 | 2019-03-26 | 青岛海尔软件有限公司 | A kind of intelligent Answer System towards vertical field |
CN105843897A (en) * | 2016-03-23 | 2016-08-10 | 青岛海尔软件有限公司 | Vertical domain-oriented intelligent question and answer system |
CN105893560A (en) * | 2016-03-31 | 2016-08-24 | 乐视控股(北京)有限公司 | Method and device for feeding effective information back to user |
CN106095956A (en) * | 2016-06-15 | 2016-11-09 | 北京智能管家科技有限公司 | support information fission querying method and device |
CN106777274B (en) * | 2016-06-16 | 2018-05-29 | 北京理工大学 | A kind of Chinese tour field knowledge mapping construction method and system |
CN106777274A (en) * | 2016-06-16 | 2017-05-31 | 北京理工大学 | A kind of Chinese tour field knowledge mapping construction method and system |
WO2018000277A1 (en) * | 2016-06-29 | 2018-01-04 | 深圳狗尾草智能科技有限公司 | Question and answer method and system, and robot |
CN106610934A (en) * | 2016-07-08 | 2017-05-03 | 四川用联信息技术有限公司 | Novel semantic similarity solving method in intelligent manufacturing industry |
CN106227788A (en) * | 2016-07-20 | 2016-12-14 | 浪潮软件集团有限公司 | Database query method based on Lucene |
CN106339366B (en) * | 2016-08-08 | 2019-05-31 | 北京百度网讯科技有限公司 | The method and apparatus of demand identification based on artificial intelligence |
CN106339366A (en) * | 2016-08-08 | 2017-01-18 | 北京百度网讯科技有限公司 | Method and device for requirement identification based on artificial intelligence (AI) |
CN106326422B (en) * | 2016-08-24 | 2019-09-17 | 北京大学 | A kind of method and system of the food safety data information retrieval of knowledge based ontology |
CN106326422A (en) * | 2016-08-24 | 2017-01-11 | 北京大学 | Method and system for retrieving food security data information based on knowledge ontology |
CN106653016A (en) * | 2016-10-28 | 2017-05-10 | 上海智臻智能网络科技股份有限公司 | Intelligent interaction method and intelligent interaction device |
CN106653016B (en) * | 2016-10-28 | 2020-07-28 | 上海智臻智能网络科技股份有限公司 | Intelligent interaction method and device |
CN106776797A (en) * | 2016-11-22 | 2017-05-31 | 中国人名解放军理工大学 | A kind of knowledge Q-A system and its method of work based on ontology inference |
CN106844741A (en) * | 2017-02-13 | 2017-06-13 | 哈尔滨工业大学 | A kind of answer method towards specific area |
CN106997376A (en) * | 2017-02-28 | 2017-08-01 | 浙江大学 | The problem of one kind is based on multi-stage characteristics and answer sentence similarity calculating method |
WO2018157805A1 (en) * | 2017-03-03 | 2018-09-07 | 腾讯科技(深圳)有限公司 | Automatic questioning and answering processing method and automatic questioning and answering system |
CN106980652A (en) * | 2017-03-03 | 2017-07-25 | 竹间智能科技(上海)有限公司 | Intelligent answer method and system |
CN106980652B (en) * | 2017-03-03 | 2020-09-08 | 竹间智能科技(上海)有限公司 | Intelligent question and answer method and system |
US11086912B2 (en) | 2017-03-03 | 2021-08-10 | Tencent Technology (Shenzhen) Company Limited | Automatic questioning and answering processing method and automatic questioning and answering system |
CN106951558B (en) * | 2017-03-31 | 2020-06-12 | 广东睿盟计算机科技有限公司 | Data processing method of tax intelligent consultation platform based on deep search |
CN106951558A (en) * | 2017-03-31 | 2017-07-14 | 广东睿盟计算机科技有限公司 | A kind of data processing method of the tax intelligent consulting platform based on deep search |
CN107122421A (en) * | 2017-04-05 | 2017-09-01 | 北京大学 | Information retrieval method and device |
CN108334490A (en) * | 2017-04-07 | 2018-07-27 | 腾讯科技(深圳)有限公司 | Keyword extracting method and keyword extracting device |
CN108334490B (en) * | 2017-04-07 | 2021-05-07 | 腾讯科技(深圳)有限公司 | Keyword extraction method and keyword extraction device |
CN107193872A (en) * | 2017-04-14 | 2017-09-22 | 深圳前海微众银行股份有限公司 | Question and answer data processing method and device |
CN107256226B (en) * | 2017-04-28 | 2018-10-30 | 北京神州泰岳软件股份有限公司 | A kind of construction method and device of knowledge base |
CN107256227A (en) * | 2017-04-28 | 2017-10-17 | 北京神州泰岳软件股份有限公司 | Towards the semantic concept spread generating method and device of knowledge content |
CN107256226A (en) * | 2017-04-28 | 2017-10-17 | 北京神州泰岳软件股份有限公司 | The construction method and device of a kind of knowledge base |
CN107329967B (en) * | 2017-05-12 | 2019-09-24 | 北京邮电大学 | Question answering system and method based on deep learning |
CN107329967A (en) * | 2017-05-12 | 2017-11-07 | 北京邮电大学 | Question answering system and method based on deep learning |
CN107436916A (en) * | 2017-06-15 | 2017-12-05 | 百度在线网络技术(北京)有限公司 | The method and device of intelligent prompt answer |
CN107391614A (en) * | 2017-07-04 | 2017-11-24 | 重庆智慧思特大数据有限公司 | A kind of Chinese question and answer matching process based on WMD |
CN107391706A (en) * | 2017-07-28 | 2017-11-24 | 湖北文理学院 | A kind of city tour's question answering system based on mobile Internet |
CN107391706B (en) * | 2017-07-28 | 2020-06-23 | 湖北文理学院 | Urban tourism question-answering system based on mobile internet |
CN107562789A (en) * | 2017-07-28 | 2018-01-09 | 深圳前海微众银行股份有限公司 | Knowledge base problem update method, customer service robot and readable storage medium storing program for executing |
CN107679231A (en) * | 2017-10-24 | 2018-02-09 | 济南浪潮高新科技投资发展有限公司 | A kind of vertical field and the implementation method of Opening field mixed type intelligent Answer System |
CN107992518A (en) * | 2017-10-31 | 2018-05-04 | 厦门快商通信息技术有限公司 | A kind of automatic question-answering method and system for lifting Intelligent dialogue speed |
CN107992539A (en) * | 2017-11-24 | 2018-05-04 | 合肥博焱智能科技有限公司 | Scenic spot tour-guide inquiring system based on artificial intelligence |
CN108345640B (en) * | 2018-01-12 | 2021-10-12 | 上海大学 | Question and answer corpus construction method based on neural network semantic analysis |
CN108345640A (en) * | 2018-01-12 | 2018-07-31 | 上海大学 | A kind of question and answer building of corpus method based on neural network semantic analysis |
CN109344385A (en) * | 2018-01-30 | 2019-02-15 | 深圳壹账通智能科技有限公司 | Natural language processing method, apparatus, computer equipment and storage medium |
CN109344385B (en) * | 2018-01-30 | 2020-12-22 | 深圳壹账通智能科技有限公司 | Natural language processing method, device, computer equipment and storage medium |
CN108280218A (en) * | 2018-02-07 | 2018-07-13 | 逸途(北京)科技有限公司 | A kind of flow system based on retrieval and production mixing question and answer |
CN108416060A (en) * | 2018-03-23 | 2018-08-17 | 芜湖乐锐思信息咨询有限公司 | A kind of city tour's question answering system based on mobile interchange |
CN108509617A (en) * | 2018-04-04 | 2018-09-07 | 上海智臻智能网络科技股份有限公司 | Construction of knowledge base, intelligent answer method and device, storage medium, the terminal in knowledge based library |
CN108846292A (en) * | 2018-05-30 | 2018-11-20 | 中国联合网络通信集团有限公司 | Desensitization process method and device |
CN108846292B (en) * | 2018-05-30 | 2021-08-17 | 中国联合网络通信集团有限公司 | Desensitization rule generation method and device |
CN110737755A (en) * | 2018-07-03 | 2020-01-31 | 百度在线网络技术(北京)有限公司 | Retrieval method and device |
CN110750543A (en) * | 2018-07-05 | 2020-02-04 | 优视科技(中国)有限公司 | Man-machine interaction method, device, system, storage medium and electronic equipment |
CN109101579A (en) * | 2018-07-19 | 2018-12-28 | 深圳追科技有限公司 | customer service robot knowledge base ambiguity detection method |
CN108984778A (en) * | 2018-07-25 | 2018-12-11 | 南京瓦尔基里网络科技有限公司 | A kind of intelligent interaction automatically request-answering system and self-teaching method |
CN109460448A (en) * | 2018-08-31 | 2019-03-12 | 厦门快商通信息技术有限公司 | It is a kind of can autonomous configuration FAQ service framework |
CN109460502A (en) * | 2018-09-14 | 2019-03-12 | 广州神马移动信息科技有限公司 | Answer clustering method and its device, electronic equipment, computer-readable medium |
CN109582761A (en) * | 2018-09-21 | 2019-04-05 | 浙江师范大学 | A kind of Chinese intelligent Answer System method of the Words similarity based on the network platform |
CN109446305A (en) * | 2018-10-10 | 2019-03-08 | 长沙师范学院 | The construction method and system of intelligent tour customer service system |
CN109460452A (en) * | 2018-10-10 | 2019-03-12 | 长沙师范学院 | Intelligent customer service system based on ontology |
CN109446304A (en) * | 2018-10-10 | 2019-03-08 | 长沙师范学院 | Intelligent customer service session method and system |
CN109359302A (en) * | 2018-10-26 | 2019-02-19 | 重庆大学 | A kind of optimization method of field term vector and fusion sort method based on it |
CN109359302B (en) * | 2018-10-26 | 2023-04-18 | 重庆大学 | Optimization method of domain word vectors and fusion ordering method based on optimization method |
CN109614464A (en) * | 2018-10-31 | 2019-04-12 | 阿里巴巴集团控股有限公司 | Method and device for traffic issues identification |
CN109614464B (en) * | 2018-10-31 | 2023-10-27 | 创新先进技术有限公司 | Method and device for identifying business problems |
CN109271503A (en) * | 2018-11-06 | 2019-01-25 | 北京猎户星空科技有限公司 | Intelligent answer method, apparatus, equipment and storage medium |
CN109710746A (en) * | 2018-12-28 | 2019-05-03 | 北京猎户星空科技有限公司 | Information processing method, device and electronic equipment |
CN110188174B (en) * | 2019-04-19 | 2021-10-29 | 浙江工业大学 | Professional field FAQ intelligent question and answer method based on professional vocabulary mining |
CN110188174A (en) * | 2019-04-19 | 2019-08-30 | 浙江工业大学 | A kind of professional domain FAQ intelligent answer method excavated based on specialized vocabulary |
CN110275937A (en) * | 2019-05-14 | 2019-09-24 | 闽江学院 | A kind of construction method and device of question and answer robot |
CN110362651A (en) * | 2019-06-11 | 2019-10-22 | 华南师范大学 | Dialogue method, system, device and the storage medium that retrieval and generation combine |
CN110263051A (en) * | 2019-06-11 | 2019-09-20 | 出门问问信息科技有限公司 | Question and answer for question answering system are to update method, device, equipment and storage medium |
CN110309509A (en) * | 2019-06-28 | 2019-10-08 | 神思电子技术股份有限公司 | A kind of semantic knowledge base construction method |
CN110377700A (en) * | 2019-07-01 | 2019-10-25 | 厦门美域中央信息科技有限公司 | A kind of professional knowledge semantic retrieval system |
CN110851576A (en) * | 2019-10-16 | 2020-02-28 | 迈达斯智能(深圳)有限公司 | Question and answer processing method, device, equipment and readable medium |
CN112925915A (en) * | 2019-12-06 | 2021-06-08 | 株式会社日立制作所 | Data generation method and device in database |
CN111198940B (en) * | 2019-12-27 | 2023-01-31 | 北京百度网讯科技有限公司 | FAQ method, question-answer search system, electronic device, and storage medium |
CN111198940A (en) * | 2019-12-27 | 2020-05-26 | 北京百度网讯科技有限公司 | FAQ method, question-answer search system, electronic device, and storage medium |
CN111506721B (en) * | 2020-04-22 | 2022-10-11 | 福州大学 | Question-answering system and construction method for domain knowledge graph |
CN111506721A (en) * | 2020-04-22 | 2020-08-07 | 福州大学 | Question-answering system and construction method for domain knowledge graph |
CN111813911A (en) * | 2020-06-30 | 2020-10-23 | 神思电子技术股份有限公司 | Knowledge automatic acquisition and updating system based on user supervision feedback and working method thereof |
CN111984774B (en) * | 2020-08-11 | 2024-02-27 | 北京百度网讯科技有限公司 | Searching method, searching device, searching equipment and storage medium |
CN111984774A (en) * | 2020-08-11 | 2020-11-24 | 北京百度网讯科技有限公司 | Search method, device, equipment and storage medium |
CN111930911A (en) * | 2020-08-12 | 2020-11-13 | 杭州东方通信软件技术有限公司 | Rapid field question-answering method and device |
CN112527965A (en) * | 2020-12-18 | 2021-03-19 | 国家电网有限公司客户服务中心 | Automatic question answering implementation method and device based on combination of professional library and chatting library |
CN112749265A (en) * | 2021-01-08 | 2021-05-04 | 哈尔滨工业大学 | Intelligent question-answering system based on multiple information sources |
CN113239161A (en) * | 2021-05-08 | 2021-08-10 | 广州华多网络科技有限公司 | Multilingual customer service response method and corresponding device, equipment and medium thereof |
CN113360626B (en) * | 2021-07-02 | 2022-02-11 | 北京容联七陌科技有限公司 | Multi-scene mixed question-answer recommendation method for intelligent customer service robot |
CN113360626A (en) * | 2021-07-02 | 2021-09-07 | 北京容联七陌科技有限公司 | Multi-scene mixed question-answer recommendation method for intelligent customer service robot |
CN114757208B (en) * | 2022-06-10 | 2022-10-21 | 荣耀终端有限公司 | Question and answer matching method and device |
CN114757208A (en) * | 2022-06-10 | 2022-07-15 | 荣耀终端有限公司 | Question and answer matching method and device |
CN116578724A (en) * | 2023-07-14 | 2023-08-11 | 杭州朗目达信息科技有限公司 | Knowledge base knowledge structure construction method and device, storage medium and terminal |
CN116578724B (en) * | 2023-07-14 | 2023-09-29 | 杭州朗目达信息科技有限公司 | Knowledge base knowledge structure construction method and device, storage medium and terminal |
CN117132392A (en) * | 2023-10-23 | 2023-11-28 | 蓝色火焰科技成都有限公司 | Vehicle loan fraud risk early warning method and system |
CN117132392B (en) * | 2023-10-23 | 2024-01-30 | 蓝色火焰科技成都有限公司 | Vehicle loan fraud risk early warning method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101373532A (en) | FAQ Chinese request-answering system implementing method in tourism field | |
CN108804521B (en) | Knowledge graph-based question-answering method and agricultural encyclopedia question-answering system | |
US10242049B2 (en) | Method, system and storage medium for implementing intelligent question answering | |
CN107818085B (en) | Answer selection method and system for reading understanding of reading robot | |
KR100533810B1 (en) | Semi-Automatic Construction Method for Knowledge of Encyclopedia Question Answering System | |
CN103136352B (en) | Text retrieval system based on double-deck semantic analysis | |
CN1924858B (en) | Method and device for fetching new words and input method system | |
CN110555153A (en) | Question-answering system based on domain knowledge graph and construction method thereof | |
CN111460787A (en) | Topic extraction method and device, terminal device and storage medium | |
CN109471949B (en) | Semi-automatic construction method of pet knowledge graph | |
Mills et al. | Graph-based methods for natural language processing and understanding—A survey and analysis | |
CN107679035B (en) | Information intention detection method, device, equipment and storage medium | |
Wang et al. | Neural related work summarization with a joint context-driven attention mechanism | |
Vel | Pre-processing techniques of text mining using computational linguistics and python libraries | |
CN109522396B (en) | Knowledge processing method and system for national defense science and technology field | |
CN111897914A (en) | Entity information extraction and knowledge graph construction method for field of comprehensive pipe gallery | |
CN102654873A (en) | Tourism information extraction and aggregation method based on Chinese word segmentation | |
CN109033272A (en) | A kind of knowledge automatic correlation method and device based on concept | |
Wu et al. | Text categorization using automatically acquired domain ontology | |
Shrawankar et al. | Construction of news headline from detailed news article | |
Pasca et al. | Answer mining from on-line documents | |
Algosaibi et al. | Using the semantics inherent in sitemaps to learn ontologies | |
Guerram et al. | A domain independent approach for ontology semantic enrichment | |
Bruggmann et al. | Spatializing a digital text archive about history | |
KR100599450B1 (en) | Indexing system and method for question answer system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C12 | Rejection of a patent application after its publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20090225 |