CN111966781A - Data query interaction method and device, electronic equipment and storage medium - Google Patents
Data query interaction method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN111966781A CN111966781A CN202010598737.3A CN202010598737A CN111966781A CN 111966781 A CN111966781 A CN 111966781A CN 202010598737 A CN202010598737 A CN 202010598737A CN 111966781 A CN111966781 A CN 111966781A
- Authority
- CN
- China
- Prior art keywords
- query
- word
- text information
- entity
- intention
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 230000003993 interaction Effects 0.000 title claims abstract description 34
- 238000004458 analytical method Methods 0.000 claims abstract description 12
- 230000002452 interceptive effect Effects 0.000 claims description 31
- 230000015654 memory Effects 0.000 claims description 26
- 239000013598 vector Substances 0.000 claims description 17
- 239000007787 solid Substances 0.000 claims description 7
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 6
- 230000002457 bidirectional effect Effects 0.000 claims description 5
- 239000000126 substance Substances 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 abstract description 11
- 238000003058 natural language processing Methods 0.000 abstract description 5
- 238000013135 deep learning Methods 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 6
- 238000004891 communication Methods 0.000 description 4
- 238000004590 computer program Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000013136 deep learning model Methods 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000002571 modificatory effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/3332—Query translation
- G06F16/3334—Selection or weighting of terms from queries, including natural language queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/338—Presentation of query results
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
- G06F40/295—Named entity recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Human Computer Interaction (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application discloses an interaction method and device for data query, electronic equipment and a storage medium, and relates to the fields of natural language processing technology, deep learning technology and data interaction. The specific implementation scheme is as follows: acquiring a query instruction, and extracting text information corresponding to the query instruction; analyzing the text information to obtain at least one entity word in the text information; performing intention analysis according to the at least one entity word to obtain a target query intention of the text information and a plurality of query templates corresponding to the target query intention; matching the text information with the plurality of query templates to obtain the target query template matched with the text information and a target query rule corresponding to the target query template; and querying by using the text information and the target query rule to obtain a query result corresponding to the query instruction, thereby effectively improving the matching degree between the query result and the query requirement.
Description
Technical Field
The application relates to a big data processing technology in the field of artificial intelligence, in particular to a natural language processing technology, a deep learning technology and data interaction, and specifically relates to an interaction method and device for data query, electronic equipment and a storage medium.
Background
The semantic interaction is mainly to analyze and accurately understand the short texts capable of expressing the search intentions of the user through a natural Language processing technology NLP (Natural Language processing), and further feed back the information desired by the user.
The semantic interaction system is widely applied to intelligent question-answering and dialogue systems, but can only be applied to a specific knowledge base, has poor generalization capability and is easy to have the condition of' not asking questions.
Disclosure of Invention
The disclosure provides an interaction method and device for data query, electronic equipment and a storage medium.
According to a first aspect of the present disclosure, there is provided an interactive method for data query, including:
acquiring a query instruction, and extracting text information corresponding to the query instruction;
analyzing the text information to obtain at least one entity word in the text information;
performing intention analysis according to the at least one entity word to obtain a target query intention of the text information and a plurality of query templates corresponding to the target query intention;
matching the text information with the plurality of query templates to obtain the target query template matched with the text information and a target query rule corresponding to the target query template; and
and querying by using the text information and the target query rule to obtain a query result corresponding to the query instruction.
According to a second aspect of the present disclosure, there is provided an interaction apparatus for data query, including:
the acquisition module is used for acquiring a query instruction and extracting text information corresponding to the query instruction;
the first analysis module is used for analyzing the text information to obtain at least one entity word in the text information;
the second analysis module is used for performing intention analysis according to the at least one entity word so as to obtain a target query intention of the text information and a plurality of query templates corresponding to the target query intention;
the matching module is used for matching the text information with the plurality of query templates to obtain the target query template matched with the text information and a target query rule corresponding to the target query template; and
and the query module is used for querying by utilizing the text information and the target query rule so as to obtain a query result corresponding to the query instruction.
According to a third aspect of the present disclosure, there is provided an electronic apparatus, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of interacting with data queries as described in the first aspect above.
According to a fourth aspect of the present disclosure, there is provided a non-transitory computer readable storage medium storing computer instructions, wherein the computer instructions are configured to cause the computer to perform the interaction method for data query of the first aspect.
According to the technology of the application, the problem that' answer questions are easy to appear in semantic interaction is solved, and the matching degree between the query result and the query requirement is effectively improved.
It should be understood that the statements in this section do not necessarily identify key or critical features of the embodiments of the present disclosure, nor do they limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description.
Drawings
The drawings are included to provide a better understanding of the present solution and are not intended to limit the present application. Wherein:
FIG. 1 is a flowchart illustrating an interactive method for data query according to an embodiment of the present disclosure;
FIG. 2 is a flowchart of another interaction method for data query according to an embodiment of the present application;
FIG. 3 is a flowchart of another interaction method for data query according to an embodiment of the present application;
FIG. 4 is a flowchart illustrating a further method for interacting with data queries according to an embodiment of the present application;
FIG. 5 is a flowchart illustrating a further method for interacting with a data query according to an embodiment of the present application;
FIG. 6 is a flowchart illustrating a further method for interacting with data queries according to an embodiment of the present application;
FIG. 7 is a flowchart illustrating a further method for interacting with a data query according to an embodiment of the present application;
FIG. 8 is a flowchart illustrating a further method for interacting with a data query according to an embodiment of the present application;
FIG. 9 is a block diagram illustrating an interactive apparatus for data query according to an embodiment of the present application;
FIG. 10 is a block diagram of an electronic device for implementing an interactive method of data querying in an embodiment of the present application.
Detailed Description
The following description of the exemplary embodiments of the present application, taken in conjunction with the accompanying drawings, includes various details of the embodiments of the application for the understanding of the same, which are to be considered exemplary only. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present application. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
The following describes interaction methods and apparatuses for data query, electronic devices, and storage media according to embodiments of the present application with reference to the drawings.
Fig. 1 is a flowchart of an interaction method for data query according to an embodiment of the present application. It should be noted that, in the interaction device for executing main data query of the data query interaction method of the embodiment, the interaction device for data query may be an electronic device with a human-Computer interaction function, such as a PC (Personal Computer), a tablet Computer, a palmtop Computer, or a mobile terminal, and is not limited herein, or software in other hardware devices with a human-Computer interaction function.
As shown in fig. 1, the interactive method for data query in the embodiment of the present application includes the following steps:
in step 101, a query instruction is obtained, and text information corresponding to the query instruction is extracted.
It should be noted that the query instruction can be obtained through an interactive interface for data query, when the interactive interface is a voice interactive interface, the query instruction input by the user through voice can be obtained first, text information corresponding to the query instruction is extracted by performing voice conversion on the voice information, and when the interactive interface is a text interactive interface, the text information corresponding to the query instruction can be extracted from the text information input by the user directly.
In step 102, the text information is parsed to obtain at least one entity word in the text information.
It should be noted that, there may be many components composing a sentence, such as nouns, auxiliary words, dummy words, prepositions, etc., but some vocabularies do not affect the query content during query, and there is only meaning that the query sentence input by the user is complete, such as "yes", so that the text information may be parsed to obtain the entity words in the text information that are limiting to the query, where the entity words may be characters, places, organizations, time, frequency, etc.
In step 103, performing intent analysis according to at least one entity word to obtain a target query intent of the text information and a plurality of query templates corresponding to the target query intent.
The method and the device are mainly applied to query of the associated information, the query intention can be mainly the query intention of the associated relation, such as people searching, time searching, place searching and the like, and the complexity can also comprise relation searching, article searching, relativity relation searching, classmate relation searching, case searching, news searching and the like.
Under the same query intention, because the query is different in limiting conditions, for example, the query is also the number of vehicles, wherein the query statement for limiting time may be "query the number of vehicles in 6 months and 21 days", and the statement for limiting location may be "query the number of vehicles reaching beijing from tianjin", after the query intention is determined, the query template needs to be further acquired to determine what limiting conditions to specifically query based on.
In step 104, the text information is matched with a plurality of query templates to obtain a target query template matched with the text information and a target query rule corresponding to the target query template.
As described above, since the restriction conditions of the query content are different, the finally adopted query rules are also different, and therefore, after the query template is determined, the target query rule corresponding to the template needs to be obtained, so as to query the intent according to the restriction conditions selected by the user according to the query rule.
The query rules can be interfaces, operators and the like which are called during query, each target query template corresponds to one target query rule, a database which can be queried to obtain target results can be called according to the target query rules, and the query results are screened, processed and the like through the target rules, so that the process of reasoning and calculation during feedback of the query results is avoided, and the generalization capability of data query is effectively improved.
In step 105, query is performed by using the text information and the target query rule to obtain a query result corresponding to the query instruction.
That is to say, the query intention of the user is obtained by analyzing the intention of the entity words, the appropriate query template is searched according to the query intention of the user, and the query template is utilized to query the content queried by the user, so that the queried result can better meet the requirement of the user, the matching degree between the query result and the query requirement is effectively improved, and the problem of question answering is avoided.
To further clarify the above embodiment, as shown in fig. 2, the parsing step 102 of the text message to obtain at least one entity word in the text message may include the following steps:
in step 201, the words in the text information are vector-converted to obtain a first vector corresponding to each word.
In step 202, a first vector is encoded by using a two-way long-short memory network model to obtain entity words.
The entity words comprise a plurality of words, and the words are adjacent and have semantic dependency relations.
It should be noted that the bidirectional long and short memory network model is a deep learning model, and the model can better capture semantic dependency of words from front to back and from back to front. Before the two-way long-short memory network model is used, model training can be carried out through corpora in an open corpus such as encyclopedia, Wikipedia and the like, so that entity words reflecting people, places, organizations, time, frequency and the like can be recognized when the model is used.
Specifically, after the text information corresponding to the query instruction is obtained, vector coding is performed on each word in the text information, the coded text information is input into a trained bidirectional long and short memory network model word by word, and the adjacent entity words with semantic dependency relationship are obtained through the model.
Further, since chinese characters usually have a word meaning, that is, represent different meanings under different context environments, such as "apple", which may represent a fruit or a technology company, in order to more accurately identify a user intention and perform a query, it is necessary to perform disambiguation on the obtained entity words so that each entity word has a unique meaning, thereby avoiding a problem of not asking a question.
Optionally, acquiring a solid dictionary, and performing disambiguation on the solid words by using the solid dictionary; and/or acquiring first attribute information for modifying the entity words in the text information, and carrying out disambiguation processing on the entity words through the first attribute information.
For example, the entity dictionary of the technology company may be preset, the entity dictionary corresponding to the technology company may further include a primary name and a secondary name, for example, "Apple" may be the primary name of the technology company, "Apple computer company," "Apple Inc," and the like, and when disambiguating, the entity word may be matched with the primary name and the secondary name in the dictionary, and the primary name corresponding to the matching result may be the entity word after disambiguation. The entity words can be matched through a matching tree algorithm.
Of course, there are also words that cannot be disambiguated by the entity word dictionary, e.g., entity words form new specific meanings in combination with the words that define them. For example, "horse cloud" is a name of a person, and there are many people called "horse cloud", so that it can be determined who the "horse cloud" queried here is according to the words defining "horse cloud", for example, if the text message is "first fumaric cloud" or "price of horse cloud", etc., it can be seen that "horse cloud" at this time refers to people with more assets.
Therefore, the entity words are obtained by deep learning of the text information and disambiguation of the entity words, the finally obtained meaning of the entity words can be simple and single as much as possible, so that the intention for query can be identified accurately and subsequently, and the query result meeting the query requirement of the user can be obtained during query. Meanwhile, the entity recognition of the deep learning model is mainly used, and the entity dictionary and the modificatory attribute disambiguation are used as auxiliary modes, so that the generalization capability and the migration capability of the entity recognition are effectively improved, and the industrial requirements are better met.
To further clearly explain the above embodiment, as shown in fig. 3, the step 103 of performing intent resolution according to at least one entity word to obtain a target query intent of the text information and a plurality of query templates corresponding to the target query intent may include the following steps:
in step 301, at least one entity word is identified to obtain a trigger word in the at least one entity word.
It should be noted that although the entity words for restricting the query result are analyzed according to the text information, not all entity words can express the query intention, for example, in the foregoing example of the train number query, only "train number" is the query intention, that is, the two sentences query the train number instead of other entity words in the sentence, but the first sentence includes the entity word "21 days in 6 months" for time restriction, and the second sentence has the entity words "tianjin" and "beijing" for vehicle origin and arrival restriction, so in order to make the query result conform to the query requirement of the user, for example, the ticket for 21 days in 6 months, or the ticket from tianjin to beijing, etc., the trigger words in the entity words need to be identified to ensure that the queried entity conforms to the query requirement of the user.
In step 302, the trigger word is matched with a plurality of query intents in the query intention list to obtain a target query intention.
The trigger word is also obtained by making a trigger word dictionary in advance, for example, places such as "hotel", "restaurant", etc. can be used as the trigger word for inquiring the geographic location.
In step 303, a plurality of query templates corresponding to the target query intention are obtained according to the target query intention.
As can be seen from the foregoing examples of the vehicle number query, even though the query intentions are the same, the query content and the result are necessarily different based on the query constraint, and therefore, it is further necessary to further obtain a plurality of templates corresponding to the target query intention after obtaining the target query intention, so as to subsequently query the result meeting the query requirement of the user according to the corresponding query templates.
Further, as shown in fig. 4, the step 302 of matching the trigger with a plurality of query intents in the query intention list to obtain the target query intention includes:
in step 401, a trigger word is identified to obtain first type information of the trigger word.
The first type of information may be a type attribute of the trigger, and includes, for example, a place, a time, an event, and the like, for example, if "a nearest hotel" is looked up, the trigger is "hotel", if "the first type of information of the trigger is" place ", and if" a past year call record "is looked up, the trigger is" call ", and if" the first type of information of the trigger is "event", and the like.
In step 402, second type information corresponding to each query intention is obtained.
Since the query is intended to have a corresponding relationship with the trigger word, the second type information may be the same as or similar to the first type information of the trigger word in the method for acquiring the first type information, or the second type information having a corresponding relationship with the first type information is provided.
In step 403, query intents corresponding to second type information identical to the first type information are taken as candidate query intents.
In step 404, the textual information is similarity matched to the first sentence of the candidate query intent.
Here, the candidate intention is an intention belonging to the same type as the trigger, and a plurality of intentions may belong to the same type, for example, "hotel" and "restaurant" may both belong to the location trigger, that is, "hotel" and "restaurant" may both be recognized as candidate triggers when the trigger belongs to the location trigger. Therefore, in order to further accurately identify the intention of the user, similarity matching is also performed according to the first example sentence of each query intention to acquire the target query intention. Wherein, the first sentence at least comprises the characteristics of the trigger word.
Optionally, when the similarity between the text information and the first example sentence of the candidate query intention is calculated, the text information and the first example sentence may be segmented, then a corresponding vector is obtained for each segmented word, then all vectors are added and averaged to obtain a sentence vector of the text information and the first example sentence, then a cosine value between the sentence vector of the text information and the sentence vector of the first example sentence is calculated, and the similarity between the text information and the first example sentence is determined according to the cosine value.
In step 405, the candidate query intention with the highest similarity to the text information is used as the target query intention.
For example, a user inputs a query text "query a five-star hotel within 500 m", performs an entity word query "500 m", "five stars" and "hotel" on the sentence, and can determine that "hotel" is a trigger word in the sentence through an address location trigger word dictionary, that is, the query of the user is "location", so that a plurality of target query intentions expressing "location" can be used as candidate query intentions, such as "hotel", "restaurant", "scenic spot", and the like, and then are compared with first examples of query intentions of "hotel", "restaurant", "scenic spot", respectively, where the first examples may include "hotel search", "scenic spot", and the like, and at this time, it can be ascertained that the similarity between the example "hotel search" and the text information of the user is the highest, and therefore, the "hotel" can be used as the target query intention.
Therefore, the query intention of the user can be accurately obtained, and the problem that the final query result is asked for answers due to deviation in the query process is avoided.
To further clarify the above embodiment, as shown in fig. 5, the step 104 of matching the text information with a plurality of query templates to obtain a target query template matching the text information and a target query rule corresponding to the target query template includes:
in step 501, a first word slot sequence is obtained for each of a plurality of query templates.
In step 502, a second sequence of word slots of the text information is obtained.
Wherein, the word slot can be characteristic data representing the attribute and/or type of the entity word.
In step 503, the query template corresponding to the first word slot sequence that is the same as the second word slot sequence is used as a candidate query template.
In step 504, the textual information is similarity matched with a second example sentence of the candidate query template.
In step 505, the candidate query template with the highest similarity to the text information is used as the target query template.
As shown in fig. 6, the step 501 of respectively obtaining the first word slot sequence of each query template in the plurality of query templates includes:
in step 601, a second example sentence of each query template is obtained.
In step 602, the second example sentence is parsed to obtain at least one entity word and at least one keyword in the second example sentence.
In step 603, attribute recognition is performed on each entity word, and the attribute is used as a first word slot of the entity word.
In step 604, attribute recognition is performed on each keyword, and a second word slot is constructed according to the attributes of the keywords.
In step 605, the first word slot and the second word slot are constructed into a first word slot sequence according to the order of the at least one entity word and the at least one keyword in the second example sentence.
As shown in fig. 7, the step 701 of respectively obtaining the first word slot sequence of each query template in the plurality of query templates includes:
in step 701, the text information is parsed to obtain at least one keyword in the text information.
In step 702, attribute recognition is performed on each entity word in the text information, and the attribute is used as a third word slot of the entity word.
In step 703, attribute recognition is performed on each keyword, and a fourth word slot is constructed according to the attributes of the keywords.
In step 704, a second word slot sequence is constructed from the third word slot and the fourth word slot in the order of the at least one entity word and the at least one keyword in the text message.
For example, the text message is "a person who has had a call record with the household registration of kansu in 2019", the entity words "2019", "kansu", "call", and the noun "household registration" and period template "without the type information can be obtained by entity recognition, wherein the person" the household registration "and the period template" without the type information can be regarded as the keyword in the text message. Then, performing attribute identification on the inquired entity words to obtain: the keyword is subjected to attribute identification to obtain "2019" ═ TIME }, "gansu" ═ SITE }, "zhang san" ═ PERSON } and "call" ═ intent }: the household registration is { KW _ property } and the person is { KW _ pattern }, the recognition results are arranged according to the sequence of the text information, and the corresponding word slot sequence is obtained as follows:
{TIME}{.*}{KW_property}{SITE}{PERSON}{INTENTION}{KW_pattern}。
similarly, the second example sentence of each query template in the plurality of query templates can be analyzed according to the rule to obtain the first word slot sequence of each query template.
After the first word slot sequence and the second word slot sequence are obtained, a first word slot sequence which is the same as the word slot and the word slot sequence contained in the first word slot sequence is selected from the plurality of word slot sequences to serve as a candidate query template, and optionally, a first word slot sequence which is the same as the second word slot sequence can be selected from the plurality of query templates by adopting a matching number algorithm.
Of course, even if the word slot sequences are the same, there will be a gap in detail, for example, the time of "year" for query, "the time of" year "for the query is" year "between 2019 and the person with the call record in zhang san of the family registered in kansu, the query granularity is large, and the matched candidate query template may include the query time of" year "," month "," day ", and the like, and therefore, it is further necessary to match the text information with the second example sentence of the candidate query template to obtain a query template with a higher approximation for query.
As a possible embodiment, the second example sentence of each query template may be multiple, and then the example sentence similarity in each template is weighted and averaged, so as to obtain the target query template with poor matching degree with the text information.
Optionally, the first example sentence and the second example sentence have the same content, and the difference is different from the result obtained by the same example sentence at different stages.
Therefore, the length of the query statement can be effectively increased through the word slot sequence, and therefore complex query operation is carried out.
In some embodiments, as shown in fig. 8, the querying in step 105 by using the text information and the target query rule to obtain the query result corresponding to the query instruction includes:
in step 801, the operators in the target query rule are updated by using entity words and/or keywords to obtain an updated target query rule.
In step 802, query is performed using the updated target query rule to obtain a query result meeting the query instruction.
It should be noted that the query rule is a rule for performing query according to a certain limiting condition and a query sequence of each limiting condition, and the limiting condition is usually obtained through text information, for example, a time condition of "a person who has a call record with kangsu in kansu in 2019" is [20190101, 20191231], that is, an operator in the query rule is updated according to an entity word and a keyword in the text information, so that the updated query rule meets a query requirement of a user, and then the updated query rule is used for performing query, so as to obtain a query result meeting a query instruction.
In summary, the text is analyzed to obtain the entity words, so that the entity words are analyzed to obtain the query intentions of the user, a suitable query template is searched for according to the query intentions of the user, the query template is utilized to query the content queried by the user, the queried result can better meet the requirements of the user, and the problem of unquestioning is effectively avoided.
In order to achieve the above object, the present application further provides an interaction device for data query.
Fig. 9 is a block diagram illustrating an interaction device for data query according to an embodiment of the present application. As shown in fig. 9, the interactive device 10 for data query includes:
the acquisition module 11 is configured to acquire a query instruction and extract text information corresponding to the query instruction;
the first analysis module 12 is configured to analyze text information to obtain at least one entity word in the text information;
a second parsing module 13, configured to perform intent parsing according to the at least one entity word, so as to obtain a target query intent of the text information and multiple query templates corresponding to the target query intent;
a matching module 14, configured to match the text information with the multiple query templates to obtain a target query template matched with the text information and a target query rule corresponding to the target query template; and
and the query module 15 is configured to perform query by using the text information and the target query rule to obtain a query result corresponding to the query instruction.
In some embodiments, the first parsing module 12 includes:
the first conversion sub-module is used for carrying out vector conversion on the words in the text information to obtain a first vector corresponding to each word; and
the first coding submodule is used for coding the first vector by using a bidirectional long-short memory network model so as to obtain the entity word, wherein the entity word comprises a plurality of characters, and the characters are adjacent and have a semantic dependency relationship.
In some embodiments, the first parsing module 20 further includes:
the first disambiguation submodule is used for acquiring an entity dictionary and carrying out disambiguation on the entity words by using the entity dictionary; and/or
And the second disambiguation submodule is used for acquiring first attribute information for modifying the entity words in the text information and carrying out disambiguation on the entity words through the first attribute information.
In some embodiments, the second parsing module 30 includes:
the first identification submodule is used for identifying the at least one entity word to obtain a trigger word in the at least one entity word;
the first matching sub-module is used for matching the trigger word with a plurality of query intents in a query intention list so as to obtain the target query intention; and
and the first obtaining sub-module is used for obtaining a plurality of query templates corresponding to the target query intention according to the target query intention.
In some embodiments, the first matching sub-module, specifically for,
identifying the trigger word to acquire first type information of the trigger word;
acquiring second type information corresponding to each query intention;
taking the query intention corresponding to the second type information which is the same as the first type information as a candidate query intention;
carrying out similarity matching on the text information and the first example sentence of the candidate query intention; and
and taking the candidate query intention with the highest similarity to the text information as the target query intention.
In some embodiments, matching module 14, includes:
the second obtaining sub-module is used for respectively obtaining the first word slot sequence of each query template in the plurality of query templates;
the third obtaining submodule is used for obtaining a second word slot sequence of the text information;
a first determining sub-module, configured to use the query template corresponding to the first word slot sequence that is the same as the second word slot sequence as a candidate query template;
the second matching submodule is used for carrying out similarity matching on the text information and a second example sentence of the candidate query template; and
and the second determining submodule is used for taking the candidate query template with the highest similarity with the text information as the target query template.
In some embodiments, the second acquisition submodule, in particular for,
acquiring a second example sentence of each query template;
analyzing the second example sentence to obtain at least one entity word and at least one keyword in the second example sentence;
performing attribute identification on each entity word, and taking the attribute as a first word slot of the entity word;
performing attribute identification on each keyword, and constructing a second word slot according to the attributes of the keywords; and
and constructing the first word slot sequence by the first word slot and the second word slot according to the sequence of the at least one entity word and the at least one keyword in the second example sentence.
In some embodiments, the third acquisition submodule, in particular for,
analyzing the text information to obtain at least one keyword in the text information;
performing attribute identification on each entity word in the text information, and taking the attribute as a third word slot of the entity word; and
performing attribute identification on each keyword, and constructing a fourth word slot according to the attributes of the keywords;
and constructing the second word slot sequence by the third word slot and the fourth word slot according to the sequence of the at least one entity word and the at least one keyword in the text message.
In some embodiments, the query module 15 includes:
the updating submodule is used for updating an operator in the target query rule by using the keyword so as to obtain the updated target query rule; and
and the query submodule is used for querying by using the updated target query rule so as to obtain a query result conforming to the query instruction.
According to an embodiment of the present application, an electronic device and a readable storage medium are also provided.
Fig. 10 is a block diagram of an electronic device of an interaction method for data query according to an embodiment of the present application. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the present application that are described and/or claimed herein.
As shown in fig. 10, the electronic apparatus includes: one or more processors 1001, memory 1002, and interfaces for connecting the various components, including high-speed interfaces and low-speed interfaces. The various components are interconnected using different buses and may be mounted on a common motherboard or in other manners as desired. The processor may process instructions for execution within the electronic device, including instructions stored in or on the memory to display graphical information of a GUI on an external input/output apparatus (such as a display device coupled to the interface). In other embodiments, multiple processors and/or multiple buses may be used, along with multiple memories and multiple memories, as desired. Also, multiple electronic devices may be connected, with each device providing portions of the necessary operations (e.g., as a server array, a group of blade servers, or a multi-processor system). Fig. 10 illustrates an example of one processor 1001.
The memory 1002 is a non-transitory computer readable storage medium provided herein. Wherein the memory stores instructions executable by at least one processor to cause the at least one processor to perform the interactive method for data query provided herein. The non-transitory computer readable storage medium of the present application stores computer instructions for causing a computer to perform the interaction method of data query provided by the present application.
The memory 1002, as a non-transitory computer readable storage medium, may be used to store non-transitory software programs, non-transitory computer executable programs, and modules, such as program instructions/modules corresponding to the interaction method of data query in the embodiment of the present application (for example, the obtaining module 11, the first parsing module 12, the second parsing module 13, the matching module 14, and the query module 15 shown in fig. 9). The processor 1001 executes various functional applications of the server and data processing, i.e., an interactive method of data query in the above-described method embodiments, by running non-transitory software programs, instructions, and modules stored in the memory 1002.
The memory 1002 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to use of the electronic device of the interactive method of data inquiry, and the like. Further, the memory 1002 may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory 1002 may optionally include memory located remotely from the processor 1001, which may be connected to the electronic device of the interactive method of data querying over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The electronic device of the interactive method for data query may further include: an input device 1003 and an output device 1004. The processor 1001, the memory 1002, the input device 1003, and the output device 1004 may be connected by a bus or other means, and the bus connection is exemplified in fig. 10.
The input device 1003 may receive input numeric or character information and generate key signal inputs related to user settings and function control of the electronic apparatus for the interactive method of data query, such as an input device of a touch screen, a keypad, a mouse, a track pad, a touch pad, a pointing stick, one or more mouse buttons, a track ball, a joystick, etc. The output devices 1004 may include a display device, auxiliary lighting devices (e.g., LEDs), and tactile feedback devices (e.g., vibrating motors), among others. The display device may include, but is not limited to, a Liquid Crystal Display (LCD), a Light Emitting Diode (LED) display, and a plasma display. In some implementations, the display device can be a touch screen.
Various implementations of the systems and techniques described here can be realized in digital electronic circuitry, integrated circuitry, application specific ASICs (application specific integrated circuits), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.
These computer programs (also known as programs, software applications, or code) include machine instructions for a programmable processor, and may be implemented using high-level procedural and/or object-oriented programming languages, and/or assembly/machine languages. As used herein, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, apparatus, and/or device (e.g., magnetic discs, optical disks, memory, Programmable Logic Devices (PLDs)) used to provide machine instructions and/or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term "machine-readable signal" refers to any signal used to provide machine instructions and/or data to a programmable processor.
To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic, speech, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.
The computer system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.
According to the technical scheme of the embodiment of the application, the text is analyzed to obtain the entity words, so that the entity words are subjected to intention analysis to obtain the query intention of the user, a proper query template is searched according to the query intention of the user, and the content queried by the user is queried by utilizing the query template, so that the queried result can better meet the requirements of the user, and the problem of question answering is effectively avoided.
It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present application may be executed in parallel, sequentially, or in different orders, and the present invention is not limited thereto as long as the desired results of the technical solutions disclosed in the present application can be achieved.
The above-described embodiments should not be construed as limiting the scope of the present application. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made in accordance with design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims (20)
1. An interactive method of data query, comprising:
acquiring a query instruction, and extracting text information corresponding to the query instruction;
analyzing the text information to obtain at least one entity word in the text information;
performing intention analysis according to the at least one entity word to obtain a target query intention of the text information and a plurality of query templates corresponding to the target query intention;
matching the text information with the plurality of query templates to obtain the target query template matched with the text information and a target query rule corresponding to the target query template; and
and querying by using the text information and the target query rule to obtain a query result corresponding to the query instruction.
2. The interactive method for data query according to claim 1, wherein the parsing the text information to obtain at least one entity word in the text information includes:
performing vector conversion on the words in the text information to obtain a first vector corresponding to each word;
and encoding the first vector by utilizing a bidirectional long-short memory network model to obtain the entity word, wherein the entity word comprises a plurality of characters, and the characters are adjacent and have a semantic dependency relationship.
3. The interactive method for data query of claim 2, wherein after the obtaining the entity words, further comprising:
acquiring a solid dictionary, and carrying out disambiguation processing on the solid words by using the solid dictionary; and/or
Acquiring first attribute information for modifying the entity words in the text information, and carrying out disambiguation processing on the entity words through the first attribute information.
4. The interaction method for data query according to claim 1, wherein the performing intent resolution according to the at least one entity word to obtain a target query intent of the text information and a plurality of query templates corresponding to the target query intent includes:
identifying the at least one entity word to obtain a trigger word in the at least one entity word;
matching the trigger word with a plurality of query intents in a query intention list to obtain the target query intention; and
and acquiring a plurality of query templates corresponding to the target query intention according to the target query intention.
5. The data query interaction method of claim 4, wherein the matching the trigger word with a plurality of the query intents in a query intention list to obtain the target query intention comprises:
identifying the trigger word to acquire first type information of the trigger word;
acquiring second type information corresponding to each query intention;
taking the query intention corresponding to the second type information which is the same as the first type information as a candidate query intention;
carrying out similarity matching on the text information and the first example sentence of the candidate query intention; and
and taking the candidate query intention with the highest similarity to the text information as the target query intention.
6. The interaction method for data query as claimed in claim 1, wherein said matching said text information with said plurality of query templates to obtain said target query template matching said text information and target query rules corresponding to said target query template comprises
Respectively acquiring a first word slot sequence of each query template in the plurality of query templates;
acquiring a second word slot sequence of the text information;
taking the query template corresponding to the first word slot sequence which is the same as the second word slot sequence as a candidate query template;
performing similarity matching on the text information and a second example sentence of the candidate query template; and
and taking the candidate query template with the highest similarity to the text information as the target query template.
7. The interactive method for data query according to claim 6, wherein the obtaining the first word slot sequence of each of the plurality of query templates respectively comprises:
acquiring a second example sentence of each query template;
analyzing the second example sentence to obtain at least one entity word and at least one keyword in the second example sentence;
performing attribute identification on each entity word, and taking the attribute as a first word slot of the entity word;
performing attribute identification on each keyword, and constructing a second word slot according to the attributes of the keywords; and
and constructing the first word slot sequence by the first word slot and the second word slot according to the sequence of the at least one entity word and the at least one keyword in the second example sentence.
8. The interactive method for data query of claim 6, wherein the obtaining the second word slot sequence of the text information comprises:
analyzing the text information to obtain at least one keyword in the text information;
performing attribute identification on each entity word in the text information, and taking the attribute as a third word slot of the entity word;
performing attribute identification on each keyword, and constructing a fourth word slot according to the attributes of the keywords; and
and constructing the second word slot sequence by the third word slot and the fourth word slot according to the sequence of the at least one entity word and the at least one keyword in the text message.
9. The interaction method for data query according to claim 8, wherein the querying by using the text information and the target query rule to obtain the query result corresponding to the query instruction includes:
updating an operator in the target query rule by using the entity word and/or the keyword to obtain the updated target query rule; and
and querying by using the updated target query rule to obtain a query result meeting the query instruction.
10. An interaction device for data query, comprising:
the acquisition module is used for acquiring a query instruction and extracting text information corresponding to the query instruction;
the first analysis module is used for analyzing the text information to obtain at least one entity word in the text information;
the second analysis module is used for performing intention analysis according to the at least one entity word so as to obtain a target query intention of the text information and a plurality of query templates corresponding to the target query intention;
the matching module is used for matching the text information with the plurality of query templates to obtain the target query template matched with the text information and a target query rule corresponding to the target query template; and
and the query module is used for querying by utilizing the text information and the target query rule so as to obtain a query result corresponding to the query instruction.
11. The interactive device for data query of claim 10, wherein the first parsing module comprises:
the first conversion sub-module is used for carrying out vector conversion on the words in the text information to obtain a first vector corresponding to each word; and
the first coding submodule is used for coding the first vector by using a bidirectional long-short memory network model so as to obtain the entity word, wherein the entity word comprises a plurality of characters, and the characters are adjacent and have a semantic dependency relationship.
12. The interactive method for data query of claim 11, further comprising:
the first disambiguation submodule is used for acquiring an entity dictionary and carrying out disambiguation on the entity words by using the entity dictionary; and/or
And the second disambiguation submodule is used for acquiring first attribute information for modifying the entity words in the text information and carrying out disambiguation on the entity words through the first attribute information.
13. The interactive device for data query of claim 10, wherein the second parsing module comprises:
the first identification submodule is used for identifying the at least one entity word to obtain a trigger word in the at least one entity word;
the first matching sub-module is used for matching the trigger word with a plurality of query intents in a query intention list so as to obtain the target query intention; and
and the first obtaining sub-module is used for obtaining a plurality of query templates corresponding to the target query intention according to the target query intention.
14. The interactive device for data query according to claim 10, wherein the first matching sub-module is specifically configured to,
identifying the trigger word to acquire first type information of the trigger word;
acquiring second type information corresponding to each query intention;
taking the query intention corresponding to the second type information which is the same as the first type information as a candidate query intention;
carrying out similarity matching on the text information and the first example sentence of the candidate query intention; and
and taking the candidate query intention with the highest similarity to the text information as the target query intention.
15. The interactive device for data query of claim 10, wherein the matching module comprises:
the second obtaining sub-module is used for respectively obtaining the first word slot sequence of each query template in the plurality of query templates;
the third obtaining submodule is used for obtaining a second word slot sequence of the text information;
a first determining sub-module, configured to use the query template corresponding to the first word slot sequence that is the same as the second word slot sequence as a candidate query template;
the second matching submodule is used for carrying out similarity matching on the text information and a second example sentence of the candidate query template; and
and the second determining submodule is used for taking the candidate query template with the highest similarity with the text information as the target query template.
16. The interactive device for data query of claim 15, wherein the second obtaining sub-module is specifically configured to,
acquiring a second example sentence of each query template;
analyzing the second example sentence to obtain at least one entity word and at least one keyword in the second example sentence;
performing attribute identification on each entity word, and taking the attribute as a first word slot of the entity word;
performing attribute identification on each keyword, and constructing a second word slot according to the attributes of the keywords; and
and constructing the first word slot sequence by the first word slot and the second word slot according to the sequence of the at least one entity word and the at least one keyword in the second example sentence.
17. The interactive device for data query of claim 15, wherein the third obtaining sub-module is specifically configured to,
analyzing the text information to obtain at least one keyword in the text information;
performing attribute identification on each entity word in the text information, and taking the attribute as a third word slot of the entity word; and
performing attribute identification on each keyword, and constructing a fourth word slot according to the attributes of the keywords;
and constructing the second word slot sequence by the third word slot and the fourth word slot according to the sequence of the at least one entity word and the at least one keyword in the text message.
18. The interactive device for data query of claim 17, wherein the query module comprises:
the updating submodule is used for updating an operator in the target query rule by using the keyword so as to obtain the updated target query rule; and
and the query submodule is used for querying by using the updated target query rule so as to obtain a query result conforming to the query instruction.
19. An electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the interactive method of data querying of any one of claims 1-9.
20. A non-transitory computer readable storage medium storing computer instructions for causing a computer to perform the interaction method for data query of any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010598737.3A CN111966781B (en) | 2020-06-28 | 2020-06-28 | Interaction method and device for data query, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010598737.3A CN111966781B (en) | 2020-06-28 | 2020-06-28 | Interaction method and device for data query, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111966781A true CN111966781A (en) | 2020-11-20 |
| CN111966781B CN111966781B (en) | 2024-02-20 |
Family
ID=73360925
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010598737.3A Active CN111966781B (en) | 2020-06-28 | 2020-06-28 | Interaction method and device for data query, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111966781B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112541335A (en) * | 2020-12-24 | 2021-03-23 | 北京百度网讯科技有限公司 | Method and device for generating interpretation text, electronic equipment and storage medium |
| CN112541070A (en) * | 2020-12-25 | 2021-03-23 | 北京百度网讯科技有限公司 | Method and device for excavating slot position updating corpus, electronic equipment and storage medium |
| CN113641857A (en) * | 2021-08-13 | 2021-11-12 | 三星电子（中国）研发中心 | Visual media personalized search method and device |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170177715A1 (en) * | 2015-12-21 | 2017-06-22 | Adobe Systems Incorporated | Natural Language System Question Classifier, Semantic Representations, and Logical Form Templates |
| US20180075366A1 (en) * | 2016-09-09 | 2018-03-15 | International Business Machines Corporation | System and method for generating full questions from natural language queries |
| CN108446286A (en) * | 2017-02-16 | 2018-08-24 | 阿里巴巴集团控股有限公司 | A kind of generation method, device and the server of the answer of natural language question sentence |
| CN109063221A (en) * | 2018-11-02 | 2018-12-21 | 北京百度网讯科技有限公司 | Query intention recognition methods and device based on mixed strategy |
| US20180373692A1 (en) * | 2017-06-21 | 2018-12-27 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method for parsing query based on artificial intelligence and computer device |
| CN109285545A (en) * | 2018-10-31 | 2019-01-29 | 北京小米移动软件有限公司 | Information processing method and device |
| CN110019713A (en) * | 2017-12-07 | 2019-07-16 | 上海智臻智能网络科技股份有限公司 | Based on the data retrieval method and device, equipment and storage medium for being intended to understand |
| CN110176315A (en) * | 2019-06-05 | 2019-08-27 | 京东方科技集团股份有限公司 | Medical answering method and system, electronic equipment, computer-readable medium |
| CN110287209A (en) * | 2019-06-10 | 2019-09-27 | 北京百度网讯科技有限公司 | Question and answer processing method, device, equipment and storage medium |
| CN111008309A (en) * | 2019-12-06 | 2020-04-14 | 北京百度网讯科技有限公司 | Query method and device |
| CN111198937A (en) * | 2019-12-02 | 2020-05-26 | 泰康保险集团股份有限公司 | Dialog generation device, computer-readable storage medium, and electronic device |
| CN111309876A (en) * | 2018-12-12 | 2020-06-19 | 北京嘀嘀无限科技发展有限公司 | Service request processing method and device, electronic equipment and storage medium |
-
2020
- 2020-06-28 CN CN202010598737.3A patent/CN111966781B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170177715A1 (en) * | 2015-12-21 | 2017-06-22 | Adobe Systems Incorporated | Natural Language System Question Classifier, Semantic Representations, and Logical Form Templates |
| US20180075366A1 (en) * | 2016-09-09 | 2018-03-15 | International Business Machines Corporation | System and method for generating full questions from natural language queries |
| CN108446286A (en) * | 2017-02-16 | 2018-08-24 | 阿里巴巴集团控股有限公司 | A kind of generation method, device and the server of the answer of natural language question sentence |
| US20180373692A1 (en) * | 2017-06-21 | 2018-12-27 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method for parsing query based on artificial intelligence and computer device |
| CN110019713A (en) * | 2017-12-07 | 2019-07-16 | 上海智臻智能网络科技股份有限公司 | Based on the data retrieval method and device, equipment and storage medium for being intended to understand |
| CN109285545A (en) * | 2018-10-31 | 2019-01-29 | 北京小米移动软件有限公司 | Information processing method and device |
| CN109063221A (en) * | 2018-11-02 | 2018-12-21 | 北京百度网讯科技有限公司 | Query intention recognition methods and device based on mixed strategy |
| CN111309876A (en) * | 2018-12-12 | 2020-06-19 | 北京嘀嘀无限科技发展有限公司 | Service request processing method and device, electronic equipment and storage medium |
| CN110176315A (en) * | 2019-06-05 | 2019-08-27 | 京东方科技集团股份有限公司 | Medical answering method and system, electronic equipment, computer-readable medium |
| CN110287209A (en) * | 2019-06-10 | 2019-09-27 | 北京百度网讯科技有限公司 | Question and answer processing method, device, equipment and storage medium |
| CN111198937A (en) * | 2019-12-02 | 2020-05-26 | 泰康保险集团股份有限公司 | Dialog generation device, computer-readable storage medium, and electronic device |
| CN111008309A (en) * | 2019-12-06 | 2020-04-14 | 北京百度网讯科技有限公司 | Query method and device |
Non-Patent Citations (2)
| Title |
|---|
| DENG H J 等: "Tactical intention recognition based on multi-entity Bayesian network", SYSTEMS ENGINEERING AND ELECTRONICS * |
| 李求实;王秋月;王珊;: "XML关键词检索的查询理解", 软件学报, no. 08 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112541335A (en) * | 2020-12-24 | 2021-03-23 | 北京百度网讯科技有限公司 | Method and device for generating interpretation text, electronic equipment and storage medium |
| CN112541335B (en) * | 2020-12-24 | 2023-09-01 | 北京百度网讯科技有限公司 | Method, device, electronic equipment and storage medium for generating interpretation text |
| CN112541070A (en) * | 2020-12-25 | 2021-03-23 | 北京百度网讯科技有限公司 | Method and device for excavating slot position updating corpus, electronic equipment and storage medium |
| CN112541070B (en) * | 2020-12-25 | 2024-03-22 | 北京百度网讯科技有限公司 | Mining method and device for slot updating corpus, electronic equipment and storage medium |
| CN113641857A (en) * | 2021-08-13 | 2021-11-12 | 三星电子（中国）研发中心 | Visual media personalized search method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111966781B (en) | 2024-02-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11334635B2 (en) | Domain specific natural language understanding of customer intent in self-help | |
| US10558754B2 (en) | Method and system for automating training of named entity recognition in natural language processing | |
| CN112507715A (en) | Method, device, equipment and storage medium for determining incidence relation between entities | |
| KR102491172B1 (en) | Natural language question-answering system and learning method | |
| EP4109295A1 (en) | Knowledge graph-based question answering method and apparatus, computer device, and medium | |
| US20210200813A1 (en) | Human-machine interaction method, electronic device, and storage medium | |
| CN112528001B (en) | Information query method and device and electronic equipment | |
| CN111966781A (en) | Data query interaction method and device, electronic equipment and storage medium | |
| US11907671B2 (en) | Role labeling method, electronic device and storage medium | |
| JP7397903B2 (en) | Intelligent interaction methods, devices, electronic devices and storage media | |
| CN111563385A (en) | Semantic processing method, semantic processing device, electronic equipment and media | |
| CN110569370B (en) | Knowledge graph construction method and device, electronic equipment and storage medium | |
| US11704326B2 (en) | Generalization processing method, apparatus, device and computer storage medium | |
| CN113657100A (en) | Entity identification method and device, electronic equipment and storage medium | |
| CN111984774A (en) | Search method, device, equipment and storage medium | |
| CN113360751A (en) | Intention recognition method, apparatus, device and medium | |
| CN112560425B (en) | Template generation method and device, electronic equipment and storage medium | |
| CN113609847A (en) | Information extraction method and device, electronic equipment and storage medium | |
| CN114153948A (en) | Question-answer knowledge base construction method, intelligent interaction method and device | |
| CN112015989A (en) | Method and device for pushing information | |
| CN112329429A (en) | Text similarity learning method, device, equipment and storage medium | |
| CN111832313A (en) | Method, device, equipment and medium for generating emotion collocation set in text | |
| CN110083817A (en) | A kind of name row discrimination method, apparatus, computer readable storage medium | |
| CN116244432B (en) | Pre-training method and device for language model and electronic equipment | |
| CN114036373B (en) | Searching method and device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |