CN106156365B - A kind of generation method and device of knowledge mapping - Google Patents
A kind of generation method and device of knowledge mapping Download PDFInfo
- Publication number
- CN106156365B CN106156365B CN201610628591.6A CN201610628591A CN106156365B CN 106156365 B CN106156365 B CN 106156365B CN 201610628591 A CN201610628591 A CN 201610628591A CN 106156365 B CN106156365 B CN 106156365B
- Authority
- CN
- China
- Prior art keywords
- data
- urtext
- knowledge
- text data
- knowledge mapping
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/253—Grammatical analysis; Style critique
-
- 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
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Abstract
The present invention provides a kind of generation method of knowledge mapping and devices, this method comprises: the urtext data to designated field carry out morphology, grammer and/or semantic analysis, obtain standardized text data；Factural information is extracted from the standardized text data, the factural information includes following element: entity, attribute, the relationship between entity and the relationship between entity and attribute；Structured representation is carried out to the factural information using the default form of expression, obtains the structural data pair of the factural information；Using the structural data to as knowledge entry, knowledge mapping is constructed.The generation method of knowledge mapping proposed by the present invention, can construct has targetedly knowledge mapping, meets designated field, such as children field, intelligent interaction demand, promote the interactive experience of different demands user.
Description
Technical field
The present invention relates to field of computer technology more particularly to a kind of knowledge mappings for children field intelligent interaction
Generation method and device.
Background technique
Children are to be easiest to the crowd received to Intelligent hardware currently on the market, and intelligence is mainly reflected in interactive intelligence
On, but it is seldom for childrenese and knowledge processing and research.The mode of common interactive dialogue majority retrieval, construction one are asked
One corpus answered, calculates the similarity of customer problem and corpus problem, and then provides corresponding reply, this to belong to shallow-layer friendship
Mutually.
Depth interaction needs to construct knowledge mapping to carry out knowledge excavation and reasoning.Knowledge mapping refers to entity, concept
As node, using semantic relation as the semantic network on side.Knowledge mapping makes knowledge acquisition more direct, therefore can be reading
The knowledge of semantic association is provided, to realize the facilitation of reading, intelligence and hommization.
In realizing process of the present invention, inventor is had found in the prior art the prior art has at least the following problems: existing knowledge graph
Composing majority is all pervasive purpose, is lack of pertinence, is insufficient for the intelligent interaction demand in children field.
Summary of the invention
In view of the above problems, the embodiment of the present invention proposes the generation method and device of a kind of knowledge mapping, to solve
Existing knowledge mapping is lack of pertinence, and is insufficient for designated field, such as children field, intelligent interaction demand the problem of.
According to an aspect of the invention, there is provided a kind of generation method of knowledge mapping, this method comprises:
Morphology, grammer and/or semantic analysis are carried out to the urtext data of designated field, obtain standardized text number
According to；
Factural information is extracted from the standardized text data, the factural information includes following element: entity, category
The relationship between relationship and entity and attribute between property, entity；
Structured representation is carried out to the factural information using the default form of expression, obtains the structuring of the factural information
Data pair；
Using the structural data to as knowledge entry, knowledge mapping is constructed.
Optionally, the method also includes:
The urtext of designated field is obtained from resource website, audio resource, video resource and/or third-party server
Data.
Optionally, described that morphology, grammer and/or semantic analysis are carried out to urtext data, obtain standardized text number
According to, comprising:
Paragraph structure division is carried out according to the file structure of the urtext data；
Morphology, grammer and/or semantic analysis are carried out to each paragraph structure marked off, obtain standardized text data.
It is optionally, described that paragraph structure division is carried out according to the file structure of the urtext data, comprising:
The file structure that the urtext data are determined according to file structure distribution characteristics, according to the file structure pair
The urtext data carry out paragraph structure division, or
File structure classification is carried out using paragraph of the paragraph sorter model trained in advance to the urtext data,
Paragraph structure division is carried out to the urtext data according to classification results.
Optionally, the described pair of each paragraph structure marked off carries out morphology, grammer and/or semantic analysis, comprising:
If the urtext data are Chinese resource, each paragraph structure marked off is segmented, part-of-speech tagging
And phrase chunking, and remove the punctuation mark in paragraph structure；
If the urtext data are foreign language resource, stem processing, morphology are carried out to each paragraph structure marked off
Reduction and phrase chunking, and remove the punctuation mark in paragraph structure.
It is optionally, described to extract factural information from the standardized text data, comprising:
Knowledge Extraction is carried out to the standardized text data, obtains noun present in the standardized text data,
And the relationship between each noun；
The identification that factural information is carried out to the result that Knowledge Extraction obtains, obtains the factural information.
It is optionally, described that Knowledge Extraction is carried out to the standardized text data, comprising:
Extracted from the standardized text data according to the structure feature of noun of all categories the noun of respective classes with
And the relationship between each noun, or
Classified using noun classification device model trained in advance to the word in the standardized text data, according to
Classification results identify and extract the relationship between noun and each noun of all categories.
Optionally, the method also includes:
It is stored using knowledge mapping of the relational database mode to building, or
It is stored using knowledge mapping of the Hash table mode to building, or
It is stored using knowledge mapping of the indexed mode to building.
Optionally, the method also includes:
Human-computer interaction is carried out according to the knowledge mapping of building.
According to another aspect of the present invention, a kind of generating means of knowledge mapping are provided, which includes:
Pretreatment unit carries out morphology, grammer and/or semantic analysis for the urtext data to designated field, obtains
To standardized text data；
Information extracting unit, for extracting factural information from the standardized text data, the factural information includes
Following element: entity, attribute, the relationship between entity and the relationship between entity and attribute；
Information presentation unit obtains institute for carrying out structured representation to the factural information using the default form of expression
State the structural data pair of factural information；
Construction unit, for, to as knowledge entry, constructing knowledge mapping using the structural data.
Optionally, described device further include:
Acquiring unit, it is specified for being obtained from resource website, audio resource, video resource and/or third-party server
The urtext data in field.
Optionally, the pretreatment unit, comprising:
First processing module, for carrying out paragraph structure division according to the file structure of the urtext data；
Second processing module is obtained for carrying out morphology, grammer and/or semantic analysis to each paragraph structure marked off
Standardized text data.
Optionally, the information extracting unit, comprising:
Abstraction module obtains the standardized text data for carrying out Knowledge Extraction to the standardized text data
Present in relationship between noun and each noun；
Identification module, the result for obtaining to Knowledge Extraction carry out the identification of factural information, obtain the factural information.
Optionally, described device further include:
Storage unit, for being stored using relational database mode to the knowledge mapping of building, or, using Hash table
Mode stores the knowledge mapping of building, or, being stored using knowledge mapping of the indexed mode to building.
Optionally, described device further include:
Interactive unit, for carrying out human-computer interaction according to the knowledge mapping of building.
The generation method and device of knowledge mapping provided by the invention, by extracting thing from the text data of designated field
Real information is indicated factural information with the default form of expression, and uses to preset the structuring that the form of expression is indicated
Data construct knowledge mapping to as knowledge entry, and then can construct with targetedly knowledge mapping, meet specified neck
Domain, such as children field, intelligent interaction demand, promote the interactive experience of different demands user.
The above description is only an overview of the technical scheme of the present invention, in order to better understand the technical means of the present invention,
And it can be implemented in accordance with the contents of the specification, and in order to allow above and other objects of the present invention, feature and advantage can
It is clearer and more comprehensible, the followings are specific embodiments of the present invention.
Detailed description of the invention
By reading the following detailed description of the preferred embodiment, various other advantages and benefits are common for this field
Technical staff will become clear.The drawings are only for the purpose of illustrating a preferred embodiment, and is not considered as to the present invention
Limitation.And throughout the drawings, the same reference numbers will be used to refer to the same parts.In the accompanying drawings:
Fig. 1 is a kind of flow chart of the generation method for knowledge mapping that the embodiment of the present invention proposes；
Fig. 2 is a kind of flow chart of the generation method for knowledge mapping that another embodiment of the present invention proposes；
Fig. 3 is the subdivision flow chart of step S11 in a kind of generation method for knowledge mapping that the embodiment of the present invention proposes；
Fig. 4 is the subdivision flow chart of step S12 in a kind of generation method for knowledge mapping that the embodiment of the present invention proposes；
Fig. 5 is a kind of structural block diagram of the generating means for knowledge mapping that the embodiment of the present invention proposes；
Fig. 6 is a kind of structural block diagram of the generating means for knowledge mapping that another embodiment of the present invention proposes.
Specific embodiment
The embodiment of the present invention is described below in detail, examples of the embodiments are shown in the accompanying drawings, wherein from beginning to end
Same or similar label indicates same or similar element or element with the same or similar functions.Below with reference to attached
The embodiment of figure description is exemplary, and for explaining only the invention, and is not construed as limiting the claims.
Those skilled in the art of the present technique are appreciated that unless expressly stated, singular " one " used herein, " one
It is a ", " described " and "the" may also comprise plural form.It is to be further understood that being arranged used in specification of the invention
Diction " comprising " refer to that there are the feature, integer, step, operation, element and/or component, but it is not excluded that in the presence of or addition
Other one or more features, integer, step, operation, element, component and/or their group.
Those skilled in the art of the present technique are appreciated that unless otherwise defined, all terms used herein (including technology art
Language and scientific term), there is meaning identical with the general understanding of those of ordinary skill in fields of the present invention.Should also
Understand, those terms such as defined in the general dictionary, it should be understood that have in the context of the prior art
The consistent meaning of meaning, and unless otherwise will not be explained in an idealized or overly formal meaning by specific definitions.
Fig. 1 shows a kind of flow chart of the generation method of knowledge mapping of the embodiment of the present invention.Referring to Fig.1, of the invention
Embodiment propose knowledge mapping generation method specifically includes the following steps:
S11, morphology, grammer and/or semantic analysis are carried out to the urtext data of designated field, obtains standardized text
Data.
Wherein, designated field refers to the field of currently practical application scenarios, such as the children field for child intelligence interaction,
It can be specifically determined according to practical application.Morphology, grammer and/or semantic analysis refer to the urtext data to designated field
The operation such as structuring processing and word segmentation processing is carried out based on morphology, grammer and/or semantic analysis.
S12, factural information is extracted from the standardized text data, the factural information includes following element: entity,
The relationship between relationship and entity and attribute between attribute, entity.
In the present embodiment, entity refers to name entity word and event name etc.；Attribute refers to the noun of name entity modification, such as
Age, gender, character relation etc..Wherein, the relationship of entity attribute extracts what entity shared mainly by the probability for calculating co-occurrence,
The attribute word of maximum probability.Relationship between entity, on the one hand according to the co-occurrence probabilities in sentence, on the other hand according to identification
Entity attribute relationship out extracts entity relationship.
S13, structured representation is carried out to the factural information using the default form of expression, obtains the knot of the factural information
Structure data pair.
In the present embodiment, the manifestation mode that N tuple can be used, which is realized, carries out structured representation to the factural information, obtains
The structural data pair of the factural information.
In a specific example, it is illustrated by taking triple as an example.Specifically, according to knowledge excavation as a result, identification
The relationship of entity and attribute and entity attribute is exported, triple is constructed.Each factural information can be expressed as (entity,
Attribute, relationship).
S14, using the structural data to as knowledge entry, construct knowledge mapping.
The generation method of knowledge mapping provided in an embodiment of the present invention, by extracting thing from the text data of designated field
Real information is indicated factural information with the default form of expression, and uses to preset the structuring that the form of expression is indicated
Data construct knowledge mapping to as knowledge entry, and then can construct with targetedly knowledge mapping, meet specified neck
Domain, such as children field, intelligent interaction demand, promote the interactive experience of different demands user.
In an alternate embodiment of the present invention where, as shown in Fig. 2, the method also includes following before step S11
Step:
S10, the original of designated field is obtained from resource website, audio resource, video resource and/or third-party server
Text data.
It further include from resource website, audio resource, video resource and/or before step S11 in the embodiment of the present invention
The step of urtext data of designated field are obtained in tripartite's server, one of specific following manner of the step or more
Kind:
(1), the urtext data of designated field are obtained from resource website by webpage capture method.In practical application
In, web crawlers technology can be used, webpage is grabbed, to obtain the urtext data of designated field from resource website；
And/or webpage is grabbed using network packet capturing technology, to obtain the urtext number of designated field from resource website
According to.
Wherein, packet capturing refer to the data packet for sending and receiving network transmission intercepted and captured, retransmitted, being edited, unloading etc.
Operation, network packet capturing technology can be by intercepting and capturing network data.
Web crawlers is the program for automatically extracting webpage, is the important component of search engine.Illustratively, with
For carrying out webpage capture using web crawlers technology, network crawl process includes: selected seed uniform resource locator first
These seeds URL is put into URL queue to be grabbed by URL；URL to be grabbed is taken out from URL array to be captured, is parsed wait grab
The domain name system DNS for taking URL checks webpage corresponding with URL to be grabbed, and the URL that these correspondence webpages have been checked is put into
URL queue is grabbed；Analysis has grabbed the URL in URL queue, analyzes other URL wherein included, and other URL are put into
URL queue to be grabbed, hence into next circulation.It should be noted that when being grabbed in the embodiment of the present invention to webpage
Above-mentioned any one or more crawl strategy can be used to be grabbed, the invention is not limited in this regard.
(2), the urtext data of designated field are obtained from voice resource by contents extraction audio recognition method.Tool
Body, voice resource can be changed into text by speech recognition technology, obtain urtext data.
(3), the urtext data of designated field are obtained from video resource by image-recognizing method.Specifically, view
Caption information in video resource can be extracted and converted to text by image recognition technology by frequency resource, obtain urtext
Data.
(4), the urtext data of designated field are obtained by third-party server.Specifically, can by with third party
Mechanism carries out cooperative resource, obtains the new contents resources such as writer children from the server of the third-party institution.
It should be noted that the mode of the urtext data of the acquisition designated field provided in the embodiment of the present invention, only
For for example, those skilled in the art can select any one or more above-mentioned mode to carry out according to practical application request
The acquisition of urtext data, the invention is not limited in this regard.
In an alternate embodiment of the present invention where, as shown in figure 3, the step S11 in above-described embodiment further comprises
Following steps:
S111, paragraph structure division is carried out according to the file structure of the urtext data.
Wherein, paragraph structure division is carried out according to the file structure of the urtext data in the step S111,
It specifically includes: the file structure of the urtext data is determined according to file structure distribution characteristics, according to the file structure
Paragraph structure division is carried out to the urtext data, or using paragraph sorter model trained in advance to the original text
The paragraph of notebook data carries out file structure classification, carries out paragraph structure division to the urtext data according to classification results.
In order to quickly and accurately realize that the paragraph structure of urtext data divides, in the embodiment of the present invention, by will be former
Beginning text data carries out structuring, distinguishes the paragraphs such as title, text, author, time, classification, realizes urtext data
Paragraph structure divides.Specifically.Specifically, can according to file structure distribution characteristics, such as: the position of text, length, in word
Hold etc. feature, determines the file structure of the urtext data.Or a little training corpus is manually marked, according to above-mentioned spy
Sign building paragraph sorter model classifies to paragraph, using prediction result of classifying as paragraph properties.
S112, morphology, grammer and/or semantic analysis are carried out to each paragraph structure marked off, obtains standardized text number
According to.
Wherein, morphology, grammer and/or semantic analysis are carried out to each paragraph structure marked off in the step S112,
It specifically includes: if the urtext data are Chinese resource, each paragraph structure marked off being segmented, part-of-speech tagging
And phrase chunking, and remove the punctuation mark in paragraph structure；If the urtext data are foreign language resource, to division
Each paragraph structure out carries out stem processing, lemmatization and phrase chunking, and removes the punctuation mark in paragraph structure.
In order to quickly and accurately realize that the paragraph structure of urtext data divides, the embodiment of the present invention is former by judgement
The language of beginning text data carries out Chinese word segmentation, part of speech mark to Chinese resource if urtext data are Chinese resource
Note, phrase chunking etc..Specifically available Open-Source Tools carry out morphology, grammer and/or semantic analysis to Chinese.If the textual data
When according to for foreign language resource, morphology, grammer and/or semantic analysis are carried out to Chinese resource according to corresponding language tool, for example, to English
Language resource carries out stem processing, lemmatization, phrase chunking etc., refers to removal tense, word suffix and is reduced into former word.It is specific
Morphology, grammer and/or semantic analysis can be carried out to English resources with Open-Source Tools.
In an alternate embodiment of the present invention where, as shown in figure 4, in above-described embodiment step S12 from the standard
Change in text data and extract factural information, further includes steps of
S121, Knowledge Extraction is carried out to the standardized text data, obtained present in the standardized text data
Relationship between noun and each noun.
Wherein, Knowledge Extraction is carried out to the standardized text data in the step S121, specifically included: according to each
The structure feature of the noun of classification is between the noun and each noun for extracting respective classes in the standardized text data
Relationship, or classified using noun classification device model trained in advance to the word in the standardized text data, according to
Classification results identify and extract the relationship between noun and each noun of all categories.Specifically, the relationship between noun can root
It is determined according to the co-occurrence probabilities in sentence.
S122, the identification that factural information is carried out to the result that Knowledge Extraction obtains, obtain the factural information.
In order to quickly and accurately realize the Knowledge Extraction of standardized text data, the embodiment of the present invention, by having number
According to observation, to the beginning word of noun, terminate the structure feature that the features such as word, word length determines noun of all categories, and according to
The structure feature of noun of all categories from standardized text data extract respective classes noun and each noun between pass
System, and then obtain factural information.
Illustrated in greater detail is carried out by taking name as an example below:
Firstly, extracting surname word, can be extracted according to One Hundred Family Names or from existing name.
The word probability often occurred in name is counted again, if word n times occurs in text altogether, is occurred M times in name, then word can be with
Probability as name is M/N；
Finally judgement ending, generally according to length and word probability, probability is similar with second step, calculate word among name,
End up the probability occurred, in addition the limitation (the general 2-4 word of Chinese personal name) of length may recognize that name.
In addition, in another embodiment of the invention, the method for being also based on statistical model is realized, it is specific as follows:
Firstly, construction mark corpus.To pretreated text data, the name in sentence is marked；
Secondly, extracting the structure feature of noun of all categories.Available feature includes that part of speech, word length, lexeme set, are previous
A word, preceding word part of speech, the latter word, rear word part of speech etc..
Finally, modeling and prediction.For example, based on the corpus marked and the tag file extracted, training statistical model.
Trained model is loaded when prediction, and standardized text data are predicted and identified with the noun of respective classes.
In an alternate embodiment of the present invention where, the method also includes following steps: using relational database mode
The knowledge mapping of building is stored, or the knowledge mapping of building is stored using Hash table mode, or using index
Mode stores the knowledge mapping of building.
Knowledge store is to need to consider inquiry property, search efficiency, space hold etc. for subsequent knowledge application
Factor.The embodiment of the present invention is explained the storage of knowledge mapping in the present invention by taking three kinds of storage methods as an example, tool
Body is as follows:
It is stored using knowledge mapping of the relational database mode to building.The storage mode is to structural data pair
(entity, attribute, relationship) design database table completes knowledge store and inquiry according to table key assignments.
It is stored using knowledge mapping of the Hash table mode to building.The storage mode is by knowledge agent (structuring number
According to the entity of centering) it is used as key, remaining constructs the storage of hash table as value.
It is stored using knowledge mapping of the indexed mode to building, full-text index is done to knowledge (structural data to),
It constructs forward index and inverted index completes storage and inquiry.
It implements of the invention one is optional, the method also includes following steps: according to the knowledge mapping of building
Carry out human-computer interaction.
The application method of knowledge mapping is varied, usually according to the knowledge and storage format excavated and issuer
Method completes knowledge reasoning, the process of human-computer interaction.In application, need to identify the information such as the entity in problem sentence, attribute, and
It is converted into the grammer of knowledge query, the reasoning results are finally provided according to the relationship in map.
It is deposited it should be noted that any one above-mentioned can be used when storing in the embodiment of the present invention to knowledge mapping
Storage mode realizes, the invention is not limited in this regard.
Below using Snow White's children's story in children field as specific embodiment, technical solution of the present invention is carried out detailed
Thin explanation.
One, the children's story text got is pre-processed first, obtains standardized text data.
Two, according to pretreated as a result, carrying out the Knowledge Extraction i.e. extraction of factural information to standardized text data is done.
Extracting content includes the personage in story, such as Snow White, Seven Dwarfs, queen, prince；Event, such as emperor
After ask witch mirror, Snow White eats malicious apple, and Snow White is rescued.
Three, knowledge mapping constructs
The fact that Knowledge Extraction, information saved in the form of structural data pair, utilized the structural data pair
As knowledge entry, knowledge mapping is constructed, and obtained knowledge mapping is stored.
Factural information includes personage, place, time etc..Representation, such as the triple of event indicate:
(Snow White is rescued, and is sued and laboured, Seven Dwarfs)；
(Snow White is rescued, and is rescued, Snow White)；
(Snow White is rescued, place, forest log cabin)；
Four, knowledge mapping application
Children ask: who has rescued Snow White?
Firstly, carrying out proper name identification, identify name: event: Snow White is rescued.Target is to ask the people that sues and labours.
Further according to recognition result, inquires knowledge store and find (Snow White is rescued, and is sued and laboured, Seven Dwarfs).
Provide the artificial Seven Dwarfs that sue and labour.
Reply is ultimately produced, Seven Dwarfs have rescued Snow White, finishing man-machine interaction.
For embodiment of the method, for simple description, therefore, it is stated as a series of action combinations, but this field
Technical staff should be aware of, and embodiment of that present invention are not limited by the describe sequence of actions, because implementing according to the present invention
Example, some steps may be performed in other sequences or simultaneously.Secondly, those skilled in the art should also know that, specification
Described in embodiment belong to preferred embodiment, the actions involved are not necessarily necessary for embodiments of the present invention.
Fig. 5 diagrammatically illustrates the structural block diagram of the generating means of the knowledge mapping of one embodiment of the invention.Referring to figure
5, the generating means of the knowledge mapping of the embodiment of the present invention specifically include pretreatment unit 501, information extracting unit 502, information
Indicate unit 503 and construction unit 504, in which: pretreatment unit 501, for the urtext data to designated field into
Row morphology, grammer and/or semantic analysis obtain standardized text data；Information extracting unit 502 is used for from the standardization
Factural information is extracted in text data, the factural information includes following element: entity, attribute, the relationship between entity and
Relationship between entity and attribute；Information presentation unit 503, for being tied using the default form of expression to the factural information
Structureization indicates, obtains the structural data pair of the factural information；Construction unit 504, for utilizing the structural data pair
As knowledge entry, knowledge mapping is constructed.
The generating means of knowledge mapping provided in an embodiment of the present invention, information extracting unit 502 pass through from by pre-processing
Factural information is extracted in the text data of treated the designated field of unit 501, information presentation unit 503 is to preset the form of expression
Factural information is indicated, so that construction unit 504 uses the structural data being indicated with the default form of expression to work
For knowledge entry, knowledge mapping is constructed, and then can be constructed with targetedly knowledge mapping, meets designated field, such as
Virgin field, intelligent interaction demand, promoted different demands user interactive experience.
In an alternate embodiment of the present invention where, described to obtain as shown in fig. 6, described device further includes acquiring unit 500
Unit 500 is taken, for obtaining the original of designated field from resource website, audio resource, video resource and/or third-party server
Beginning text data.
Specifically, the acquiring unit 500 can obtain the urtext number of designated field by following at least one mode
According to:
The urtext data of designated field are obtained from resource website by webpage capture method；
The urtext data of designated field are obtained from voice resource by contents extraction audio recognition method；
The urtext data of designated field are obtained from video resource by image-recognizing method；
The urtext data of designated field are obtained by third-party server.
In an alternate embodiment of the present invention where, the pretreatment unit 501, including at first processing module and second
Manage module, in which: first processing module, for carrying out paragraph structure division according to the file structure of the urtext data；
Second processing module obtains standardization text for carrying out morphology, grammer and/or semantic analysis to each paragraph structure marked off
Notebook data.
Wherein, first processing module, specifically for determining the urtext data according to file structure distribution characteristics
File structure carries out paragraph structure division to the urtext data according to the file structure, or using training in advance
Paragraph sorter model carries out file structure classification to the paragraph of the urtext data, according to classification results to described original
Text data carries out paragraph structure division.
In order to quickly and accurately realize that the paragraph structure of urtext data divides, in the embodiment of the present invention, the first processing
Module is distinguished the paragraphs such as title, text, author, time, classification, is realized former by the way that urtext data are carried out structuring
The paragraph structure of beginning text data divides.Specifically.Specifically, can according to file structure distribution characteristics, such as: the position of text
It sets, length, word content etc. feature, determines the file structure of the urtext data.Or the artificial a little training of mark
Corpus constructs paragraph sorter model according to features described above and classifies to paragraph, using prediction result of classifying as paragraph properties.
Wherein, Second processing module, it is each to what is marked off if be Chinese resource specifically for the urtext data
Paragraph structure segmented, part-of-speech tagging and phrase chunking, and removes the punctuation mark in paragraph structure；If the original text
When notebook data is foreign language resource, stem processing, lemmatization and phrase chunking are carried out to each paragraph structure marked off, and go
Except the punctuation mark in paragraph structure.
In order to quickly and accurately realize that the paragraph structure of urtext data divides, the embodiment of the present invention, second processing mould
Block passes through the language for judging urtext data, if urtext data are Chinese resource, carries out Chinese to Chinese resource
Participle, part-of-speech tagging, phrase chunking etc..Specifically available Open-Source Tools carry out morphology, grammer and/or semantic analysis to Chinese.
If the text data is foreign language resource, morphology, grammer and/or semanteme point are carried out to Chinese resource according to corresponding language tool
Analysis, for example, carrying out stem processing, lemmatization, phrase chunking etc. to English resources, referring to removal tense, word suffix and being reduced into
Former word.Specifically morphology, grammer and/or semantic analysis can also be carried out to English resources with Open-Source Tools.
In an alternate embodiment of the present invention where, the information extracting unit 502, including abstraction module and identification mould
Block, in which: abstraction module obtains the standardized text data for carrying out Knowledge Extraction to the standardized text data
Present in relationship between noun and each noun；Identification module, the result for obtaining to Knowledge Extraction carry out true letter
The identification of breath obtains the factural information.
Wherein, abstraction module, specifically for according to the structure feature of noun of all categories from the standardized text data
Relationship between the middle noun and each noun for extracting respective classes, or using noun classification device model trained in advance to described
Word in standardized text data is classified, identified and extracted according to classification results noun of all categories and each noun it
Between relationship.Specifically, the relationship between noun can be determined according to the co-occurrence probabilities in sentence.
In order to quickly and accurately realize the Knowledge Extraction of standardized text data, the embodiment of the present invention, information extracting unit
502 by the observation to data with existing, to the beginning word of noun, terminates the features such as word, word length and determines noun of all categories
Structure feature, and extracted from standardized text data according to the structure feature of noun of all categories respective classes noun and
Relationship between each noun, and then obtain factural information.
In an alternate embodiment of the present invention where, described device further includes attached storage unit not shown in the figure, this is deposited
Storage unit, for being stored using relational database mode to the knowledge mapping of building, or, using Hash table mode to building
Knowledge mapping stored, or, being stored using knowledge mapping of the indexed mode to building.
In an alternate embodiment of the present invention where, described device further includes attached interactive unit not shown in the figure, the friendship
Mutual unit, for carrying out human-computer interaction according to the knowledge mapping of building.
For device embodiment, since it is basically similar to the method embodiment, related so being described relatively simple
Place illustrates referring to the part of embodiment of the method.
In conclusion the generation method and device of knowledge mapping provided in an embodiment of the present invention, by from designated field
Factural information is extracted in text data, factural information is indicated with the default form of expression, and is used to preset the form of expression
The structural data being indicated constructs knowledge mapping to as knowledge entry, and then can construct to have and targetedly know
Know map, meet designated field, such as children field, intelligent interaction demand, promote the interactive experience of different demands user.
Through the above description of the embodiments, those skilled in the art can be understood that the present invention can lead to
Hardware realization is crossed, the mode of necessary general hardware platform can also be added to realize by software.Based on this understanding, this hair
Bright technical solution can be embodied in the form of software products, which can store in a non-volatile memories
In medium (can be CD-ROM, USB flash disk, mobile hard disk etc.), including some instructions are used so that a computer equipment (can be
Personal computer, server or network equipment etc.) execute method described in each embodiment of the present invention.
It will be appreciated by those skilled in the art that attached drawing is the schematic diagram of a preferred embodiment, module or stream in attached drawing
Journey is not necessarily implemented necessary to the present invention.
It will be appreciated by those skilled in the art that the module in system in embodiment can describe be divided according to embodiment
It is distributed in the system of embodiment, corresponding change can also be carried out and be located in one or more systems different from the present embodiment.On
The module for stating embodiment can be merged into a module, can also be further split into multiple submodule.
The above is only some embodiments of the invention, it is noted that for the ordinary skill people of the art
For member, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications are also answered
It is considered as protection scope of the present invention.
Claims (10)
1. a kind of generation method of knowledge mapping characterized by comprising
Morphology, grammer and/or semantic analysis are carried out to the urtext data of designated field, obtain standardized text data；
Factural information is extracted from the standardized text data, the factural information includes following element: entity, attribute, reality
The relationship between relationship and entity and attribute between body；
Structured representation is carried out to the factural information using the default form of expression, obtains the structural data of the factural information
It is right；
Using the structural data to as knowledge entry, knowledge mapping is constructed；
The urtext data to designated field carry out morphology, grammer and/or semantic analysis, obtain standardized text data
Include:
The file structure that the urtext data are determined according to file structure distribution characteristics, according to the file structure to described
Urtext data carry out paragraph structure division, or using paragraph sorter model trained in advance to the urtext data
Paragraph carry out file structure classification, according to classification results to the urtext data carry out paragraph structure division；
Morphology, grammer and/or semantic analysis are carried out to each paragraph structure marked off, obtain standardized text data.
2. the method according to claim 1, wherein the method also includes:
The urtext number of designated field is obtained from resource website, audio resource, video resource and/or third-party server
According to.
3. method according to claim 1 or 2, which is characterized in that the described pair of each paragraph structure progress morphology marked off,
Grammer and/or semantic analysis, comprising:
If the urtext data are Chinese resource, each paragraph structure marked off is segmented, part-of-speech tagging and
Phrase chunking, and remove the punctuation mark in paragraph structure；
If the urtext data are foreign language resource, stem processing, lemmatization are carried out to each paragraph structure marked off
And phrase chunking, and remove the punctuation mark in paragraph structure.
4. method according to claim 1 or 2, which is characterized in that described to extract thing from the standardized text data
Real information, comprising:
Knowledge Extraction is carried out to the standardized text data, obtains noun present in the standardized text data, and
Relationship between each noun；
The identification that factural information is carried out to the result that Knowledge Extraction obtains, obtains the factural information.
5. according to the method described in claim 4, it is characterized in that, described carry out knowledge pumping to the standardized text data
It takes, comprising:
The noun of respective classes and each is extracted from the standardized text data according to the structure feature of noun of all categories
Relationship between noun, or
Classified using noun classification device model trained in advance to the word in the standardized text data, according to classification
As a result it identifies and extracts the relationship between noun and each noun of all categories.
6. the method according to claim 1, wherein the method also includes:
It is stored using knowledge mapping of the relational database mode to building, or
It is stored using knowledge mapping of the Hash table mode to building, or
It is stored using knowledge mapping of the indexed mode to building.
7. a kind of generating means of knowledge mapping characterized by comprising
Pretreatment unit carries out morphology, grammer and/or semantic analysis for the urtext data to designated field, is marked
Standardization text data；
Information extracting unit, for extracting factural information from the standardized text data, the factural information includes following
Element: entity, attribute, the relationship between entity and the relationship between entity and attribute；
Information presentation unit obtains the thing for carrying out structured representation to the factural information using the default form of expression
The structural data pair of real information；
Construction unit, for, to as knowledge entry, constructing knowledge mapping using the structural data；
The pretreatment unit, including first processing module and Second processing module, in which:
The first processing module, for determining the file structure of the urtext data according to file structure distribution characteristics,
Paragraph structure division is carried out to the urtext data according to the file structure, or using paragraph classifier trained in advance
Model carries out file structure classification to the paragraphs of the urtext data, according to classification results to the urtext data into
Row paragraph structure divides；
The Second processing module is obtained for carrying out morphology, grammer and/or semantic analysis to each paragraph structure marked off
Standardized text data.
8. device according to claim 7, which is characterized in that described device further include:
Acquiring unit, for obtaining designated field from resource website, audio resource, video resource and/or third-party server
Urtext data.
9. device according to claim 7 or 8, which is characterized in that the information extracting unit, comprising:
Abstraction module obtains depositing in the standardized text data for carrying out Knowledge Extraction to the standardized text data
Noun and each noun between relationship；
Identification module, the result for obtaining to Knowledge Extraction carry out the identification of factural information, obtain the factural information.
10. device according to claim 7, which is characterized in that described device further include:
Storage unit, for being stored using relational database mode to the knowledge mapping of building, or, using Hash table mode
The knowledge mapping of building is stored, or, being stored using knowledge mapping of the indexed mode to building.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610628591.6A CN106156365B (en) | 2016-08-03 | 2016-08-03 | A kind of generation method and device of knowledge mapping |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610628591.6A CN106156365B (en) | 2016-08-03 | 2016-08-03 | A kind of generation method and device of knowledge mapping |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106156365A CN106156365A (en) | 2016-11-23 |
| CN106156365B true CN106156365B (en) | 2019-06-18 |
Family
ID=57328826
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610628591.6A Active CN106156365B (en) | 2016-08-03 | 2016-08-03 | A kind of generation method and device of knowledge mapping |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106156365B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11829726B2 (en) | 2021-01-25 | 2023-11-28 | International Business Machines Corporation | Dual learning bridge between text and knowledge graph |
Families Citing this family (61)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106294321B (en) * | 2016-08-04 | 2019-05-31 | 北京儒博科技有限公司 | A kind of the dialogue method for digging and device of specific area |
| CN106599091B (en) * | 2016-11-24 | 2020-07-14 | 上海交通大学 | RDF graph structure storage and index method based on key value storage |
| CN106776564B (en) * | 2016-12-21 | 2020-04-24 | 张永成 | Semantic recognition method and system based on knowledge graph |
| CN106777331A (en) * | 2017-01-11 | 2017-05-31 | 北京航空航天大学 | Knowledge mapping generation method and device |
| CN106933804B (en) * | 2017-03-10 | 2020-03-31 | 上海数眼科技发展有限公司 | Structured information extraction method based on deep learning |
| CN106934042B (en) * | 2017-03-16 | 2020-05-29 | 中国人民解放军国防科学技术大学 | Knowledge graph representation system and implementation method thereof |
| CN107016072B (en) * | 2017-03-23 | 2020-05-15 | 成都市公安科学技术研究所 | Knowledge inference system and method based on social network knowledge graph |
| CN107122444A (en) * | 2017-04-24 | 2017-09-01 | 北京科技大学 | A kind of legal knowledge collection of illustrative plates method for auto constructing |
| CN107169078A (en) | 2017-05-10 | 2017-09-15 | 京东方科技集团股份有限公司 | Knowledge of TCM collection of illustrative plates and its method for building up and computer system |
| CN107066621B (en) * | 2017-05-11 | 2022-11-08 | 腾讯科技（深圳）有限公司 | Similar video retrieval method and device and storage medium |
| CN107341215B (en) * | 2017-06-07 | 2020-05-12 | 北京航空航天大学 | Multi-source vertical knowledge graph classification integration query system based on distributed computing platform |
| CN107301235A (en) * | 2017-06-27 | 2017-10-27 | 山东浪潮商用系统有限公司 | A kind of communicating knowledge collection of illustrative plates display systems |
| CN107526795B (en) * | 2017-08-17 | 2020-05-29 | 晶赞广告（上海）有限公司 | Knowledge base construction method and device, storage medium and computing equipment |
| CN107633075A (en) * | 2017-09-22 | 2018-01-26 | 吉林大学 | A kind of multi-source heterogeneous data fusion platform and fusion method |
| CN107908671B (en) * | 2017-10-25 | 2022-02-01 | 南京擎盾信息科技有限公司 | Knowledge graph construction method and system based on legal data |
| CN107832407B (en) * | 2017-11-03 | 2020-08-25 | 上海点融信息科技有限责任公司 | Information processing method and device for generating knowledge graph and readable storage medium |
| CN108182245A (en) * | 2017-12-28 | 2018-06-19 | 北京锐安科技有限公司 | The construction method and device of people's object properties classificating knowledge collection of illustrative plates |
| CN108133030A (en) * | 2017-12-29 | 2018-06-08 | 北京物灵智能科技有限公司 | A kind of realization method and system for painting this question and answer |
| CN108170813A (en) * | 2017-12-29 | 2018-06-15 | 智搜天机（北京）信息技术有限公司 | A kind of method and its system of full media content intelligent checks |
| CN108304493B (en) * | 2018-01-10 | 2020-06-12 | 深圳市腾讯计算机系统有限公司 | Hypernym mining method and device based on knowledge graph |
| CN108197119A (en) * | 2018-02-05 | 2018-06-22 | 成都卓观信息技术有限公司 | The archives of paper quality digitizing solution of knowledge based collection of illustrative plates |
| CN110209827B (en) * | 2018-02-07 | 2023-09-19 | 腾讯科技（深圳）有限公司 | Search method, search device, computer-readable storage medium, and computer device |
| CN108536724A (en) * | 2018-02-13 | 2018-09-14 | 西安理工大学 | Main body recognition methods in a kind of metro design code based on the double-deck hash index |
| CN108665141B (en) * | 2018-04-03 | 2022-03-22 | 山东科技大学 | Method for automatically extracting emergency response process model from emergency plan |
| JP7095377B2 (en) * | 2018-04-17 | 2022-07-05 | 富士フイルムビジネスイノベーション株式会社 | Information processing equipment and information processing programs |
| CN108874915A (en) * | 2018-05-30 | 2018-11-23 | 苏州思必驰信息科技有限公司 | Method of Knowledge Organization, system, electronic equipment and storage medium |
| CN109002435A (en) * | 2018-06-06 | 2018-12-14 | 达而观信息科技（上海）有限公司 | A kind of data processing method and device |
| CN109582799B (en) * | 2018-06-29 | 2020-09-22 | 北京百度网讯科技有限公司 | Method and device for determining knowledge sample data set and electronic equipment |
| CN110851610B (en) * | 2018-07-25 | 2022-09-27 | 百度在线网络技术（北京）有限公司 | Knowledge graph generation method and device, computer equipment and storage medium |
| CN109347798A (en) * | 2018-09-12 | 2019-02-15 | 东软集团股份有限公司 | Generation method, device, equipment and the storage medium of network security knowledge map |
| TWI682287B (en) | 2018-10-25 | 2020-01-11 | 財團法人資訊工業策進會 | Knowledge graph generating apparatus, method, and computer program product thereof |
| CN109657065A (en) * | 2018-10-31 | 2019-04-19 | 百度在线网络技术（北京）有限公司 | Knowledge mapping processing method, device and electronic equipment |
| CN109189947A (en) * | 2018-11-07 | 2019-01-11 | 曲阜师范大学 | A kind of mobile data knowledge mapping method for auto constructing based on relational database |
| CN109582958B (en) * | 2018-11-20 | 2023-07-18 | 厦门大学深圳研究院 | Disaster story line construction method and device |
| CN109523988B (en) * | 2018-11-26 | 2021-11-05 | 安徽淘云科技股份有限公司 | Text deduction method and device |
| CN111259160B (en) * | 2018-11-30 | 2023-08-29 | 百度在线网络技术（北京）有限公司 | Knowledge graph construction method, device, equipment and storage medium |
| CN109299290A (en) * | 2018-12-07 | 2019-02-01 | 广东小天才科技有限公司 | A kind of dub in background music recommended method and the electronic equipment of knowledge based map |
| CN111368145A (en) * | 2018-12-26 | 2020-07-03 | 沈阳新松机器人自动化股份有限公司 | Knowledge graph creating method and system and terminal equipment |
| CN109933671A (en) * | 2019-01-31 | 2019-06-25 | 平安科技（深圳）有限公司 | Construct method, apparatus, computer equipment and the storage medium of personal knowledge map |
| US11403328B2 (en) | 2019-03-08 | 2022-08-02 | International Business Machines Corporation | Linking and processing different knowledge graphs |
| CN110134842B (en) * | 2019-04-03 | 2021-08-31 | 深圳价值在线信息科技股份有限公司 | Information matching method and device based on information map, storage medium and server |
| CN110222198A (en) * | 2019-06-18 | 2019-09-10 | 卓尔智联(武汉)研究院有限公司 | Non-ferrous metal industry knowledge mapping construction method, electronic device and storage medium |
| CN110275965B (en) * | 2019-06-27 | 2021-12-21 | 卓尔智联(武汉)研究院有限公司 | False news detection method, electronic device and computer readable storage medium |
| US11379733B2 (en) | 2019-07-10 | 2022-07-05 | International Business Machines Corporation | Detecting and predicting object events from images |
| CN110347845A (en) * | 2019-07-15 | 2019-10-18 | 北京明略软件系统有限公司 | The method for drafting and device of knowledge mapping |
| CN110750651B (en) * | 2019-10-16 | 2023-05-26 | 同方知网数字出版技术股份有限公司 | Knowledge graph construction method based on scientific and technological achievements |
| CN110738982B (en) * | 2019-10-22 | 2022-01-28 | 珠海格力电器股份有限公司 | Request processing method and device and electronic equipment |
| CN111160841A (en) * | 2019-11-29 | 2020-05-15 | 广东轩辕网络科技股份有限公司 | Organization architecture construction method and device based on knowledge graph |
| CN111339311A (en) * | 2019-12-30 | 2020-06-26 | 智慧神州(北京)科技有限公司 | Method, device and processor for extracting structured events based on generative network |
| CN111259163A (en) * | 2020-01-14 | 2020-06-09 | 北京明略软件系统有限公司 | Knowledge graph generation method and device and computer readable storage medium |
| CN111460080B (en) * | 2020-03-25 | 2022-04-22 | 中国人民解放军国防科技大学 | Event map construction and query method and system for open source data heat analysis |
| CN112001825B (en) * | 2020-08-18 | 2021-06-01 | 上海松鼠课堂人工智能科技有限公司 | Learning cognitive path planning system based on cognitive map |
| CN112148893A (en) * | 2020-09-25 | 2020-12-29 | 南方电网数字电网研究院有限公司 | Energy analysis knowledge graph construction method and energy analysis visualization method |
| CN112613315A (en) * | 2020-12-29 | 2021-04-06 | 重庆农村商业银行股份有限公司 | Text knowledge automatic extraction method, device, equipment and storage medium |
| CN112733515B (en) * | 2020-12-31 | 2022-11-11 | 贝壳技术有限公司 | Text generation method and device, electronic equipment and readable storage medium |
| CN112765363B (en) * | 2021-01-19 | 2022-11-22 | 昆明理工大学 | Demand map construction method for scientific and technological service demand |
| CN112951446A (en) * | 2021-04-16 | 2021-06-11 | 平安科技（深圳）有限公司 | Medicine query method, device, equipment and storage medium based on medicine atlas |
| CN113220835B (en) * | 2021-05-08 | 2023-09-29 | 北京百度网讯科技有限公司 | Text information processing method, device, electronic equipment and storage medium |
| CN113609309B (en) * | 2021-08-16 | 2024-02-06 | 脸萌有限公司 | Knowledge graph construction method and device, storage medium and electronic equipment |
| CN116401375B (en) * | 2023-03-23 | 2024-02-20 | 深圳宏鹏数字供应链管理有限公司 | Knowledge graph construction method and system |
| CN116955639A (en) * | 2023-04-24 | 2023-10-27 | 浙商期货有限公司 | Method and device for constructing future industry chain knowledge graph and computer equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003196296A (en) * | 2001-12-25 | 2003-07-11 | Celestar Lexico-Sciences Inc | Document knowledge managing device, document knowledge managing method, its program, and recording medium |
| US6598043B1 (en) * | 1999-10-04 | 2003-07-22 | Jarg Corporation | Classification of information sources using graph structures |
| CN103699663A (en) * | 2013-12-27 | 2014-04-02 | 中国科学院自动化研究所 | Hot event mining method based on large-scale knowledge base |
| CN105550190A (en) * | 2015-06-26 | 2016-05-04 | 许昌学院 | Knowledge graph-oriented cross-media retrieval system |
| CN105574098A (en) * | 2015-12-11 | 2016-05-11 | 百度在线网络技术（北京）有限公司 | Knowledge graph generation method and device and entity comparing method and device |
-
2016
- 2016-08-03 CN CN201610628591.6A patent/CN106156365B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6598043B1 (en) * | 1999-10-04 | 2003-07-22 | Jarg Corporation | Classification of information sources using graph structures |
| JP2003196296A (en) * | 2001-12-25 | 2003-07-11 | Celestar Lexico-Sciences Inc | Document knowledge managing device, document knowledge managing method, its program, and recording medium |
| CN103699663A (en) * | 2013-12-27 | 2014-04-02 | 中国科学院自动化研究所 | Hot event mining method based on large-scale knowledge base |
| CN105550190A (en) * | 2015-06-26 | 2016-05-04 | 许昌学院 | Knowledge graph-oriented cross-media retrieval system |
| CN105574098A (en) * | 2015-12-11 | 2016-05-11 | 百度在线网络技术（北京）有限公司 | Knowledge graph generation method and device and entity comparing method and device |
Non-Patent Citations (2)
| Title |
|---|
| "基于Web的实体信息提取和搜索研究";周安林;《中国优秀硕士学位论文全文数据库 信息科技辑》;20160315;正文第2.32节 |
| "知识图谱构建技术综述";刘峤 等;《计算机研究与发展》;20160331;全文 |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11829726B2 (en) | 2021-01-25 | 2023-11-28 | International Business Machines Corporation | Dual learning bridge between text and knowledge graph |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106156365A (en) | 2016-11-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106156365B (en) | A kind of generation method and device of knowledge mapping | |
| US10565244B2 (en) | System and method for text categorization and sentiment analysis | |
| CN109657054B (en) | Abstract generation method, device, server and storage medium | |
| US11514235B2 (en) | Information extraction from open-ended schema-less tables | |
| US10423649B2 (en) | Natural question generation from query data using natural language processing system | |
| CN104503998B (en) | For the kind identification method and device of user query sentence | |
| CN106960030B (en) | Information pushing method and device based on artificial intelligence | |
| CN103593412B (en) | A kind of answer method and system based on tree structure problem | |
| CN109697239B (en) | Method for generating teletext information | |
| US11436446B2 (en) | Image analysis enhanced related item decision | |
| CN110287314B (en) | Long text reliability assessment method and system based on unsupervised clustering | |
| CN109992781B (en) | Text feature processing method and device and storage medium | |
| CN111506794A (en) | Rumor management method and device based on machine learning | |
| CN111046656A (en) | Text processing method and device, electronic equipment and readable storage medium | |
| CN111782793A (en) | Intelligent customer service processing method, system and equipment | |
| CN112069312A (en) | Text classification method based on entity recognition and electronic device | |
| CN111368066B (en) | Method, apparatus and computer readable storage medium for obtaining dialogue abstract | |
| JP6867963B2 (en) | Summary Evaluation device, method, program, and storage medium | |
| CN109660621A (en) | A kind of content delivery method and service equipment | |
| CN109672586A (en) | A kind of DPI service traffics recognition methods, device and computer readable storage medium | |
| CN111783425A (en) | Intention identification method based on syntactic analysis model and related device | |
| CN111199151A (en) | Data processing method and data processing device | |
| CN113094471A (en) | Interactive data processing method and device | |
| KR101126186B1 (en) | Apparatus and Method for disambiguation of morphologically ambiguous Korean verbs, and Recording medium thereof | |
| CN111680493A (en) | English text analysis method and device, readable storage medium and computer equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: Room 508-598, Xitian Gezhuang Town Government Office Building, No. 8 Xitong Road, Miyun County Economic Development Zone, Beijing 101500 Applicant after: Beijing Rubo Technology Co., Ltd. Address before: Room 508-598, Xitian Gezhuang Town Government Office Building, No. 8 Xitong Road, Miyun County Economic Development Zone, Beijing 101500 Applicant before: BEIJING INTELLIGENT HOUSEKEEPER TECHNOLOGY CO., LTD. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20210819 Address after: 301-112, floor 3, building 2, No. 18, YANGFANGDIAN Road, Haidian District, Beijing 100038 Patentee after: Beijing Rubu Technology Co.,Ltd. Address before: Room 508-598, Xitian Gezhuang Town Government Office Building, No. 8 Xitong Road, Miyun County Economic Development Zone, Beijing 101500 Patentee before: BEIJING ROOBO TECHNOLOGY Co.,Ltd. |
|
| TR01 | Transfer of patent right |