CN107895027A - Individual feelings and emotions knowledge mapping method for building up and device - Google Patents
Individual feelings and emotions knowledge mapping method for building up and device Download PDFInfo
- Publication number
- CN107895027A CN107895027A CN201711145598.3A CN201711145598A CN107895027A CN 107895027 A CN107895027 A CN 107895027A CN 201711145598 A CN201711145598 A CN 201711145598A CN 107895027 A CN107895027 A CN 107895027A
- Authority
- CN
- China
- Prior art keywords
- mrow
- emotion
- msub
- vocabulary
- extraction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- 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
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/237—Lexical tools
- G06F40/247—Thesauruses; Synonyms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Animal Behavior & Ethology (AREA)
- Machine Translation (AREA)
Abstract
The present invention relates to a kind of individual feelings and emotions knowledge mapping method for building up and device, computer equipment and computer-readable recording medium, emotion object vocabulary and emotion vocabulary are extracted in this method from the personal text conversation and/or the personal this paper questionnaires done first, default benchmark emotion vocabulary is based on again, the emotion vocabulary of extraction is sorted out, determine its Sentiment orientation, finally structure is personal, the triple structure of emotion object vocabulary and Sentiment orientation, and using the triple structure individual feelings and emotions knowledge mapping personal as this.So as to according to the substantial amounts of language material structure of the individual more completely with for accurate emotion knowledge mapping, more valuable reference is provided for the application scenarios of each emotion knowledge mapping.
Description
Technical field
The present invention relates to software technology field, and in particular to a kind of individual feelings and emotions knowledge mapping method for building up and device.
Background technology
Knowledge mapping is a figure knowledge base of relation between entity and entity, and its essence is a kind of explanation entity knowledge
Between semantic network figure.The foundation of knowledge mapping is broadly divided into：Text analyzing, structural analysis, link are integrated.At present increasingly
More researchers has carried out the research for knowledge mapping, especially for individual individual feelings and emotions knowledge mapping research,
Like so as to understand the different of the personage according to the individual feelings and emotions knowledge mapping of individual, and had according to different hobbies
Targetedly measure, such as targetedly message push etc. can be carried out by platform on the internet.
However, inventor has found during the embodiment of the present invention is implemented, the foundation side of some existing knowledge mappings
Method can only be built according to single sentence, lack context environmental feature, and the text analyzing for big section does not go out
Color, the knowledge mapping model built accurately can not comprehensively reflect the emotion of individual.
The content of the invention
The purpose of the embodiment of the present invention is to provide a kind of individual feelings and emotions knowledge mapping method for building up and device.
In a first aspect, the embodiments of the invention provide a kind of individual feelings and emotions knowledge mapping method for building up,
The personal user data of knowledge mapping to be established is obtained, the user data includes and the personal text pair
This paper questionnaires that words and/or the individual are done；
It is the feelings that emotion object vocabulary and emotion vocabulary, the emotion object vocabulary are extracted in the user data
The object that sense vocabulary is limited；
Based on default benchmark emotion vocabulary, according to the similarity of the emotion vocabulary of extraction and the benchmark emotion vocabulary,
The emotion vocabulary of the extraction is sorted out, and the emotion vocabulary using the Sentiment orientation of generic vocabulary as the extraction
Sentiment orientation；
Build the triple of the Sentiment orientation of personal, the described emotion object vocabulary and the emotion vocabulary of the extraction
Structure, using the triple structure as the personal individual feelings and emotions knowledge mapping.
Alternatively, extraction emotion object vocabulary and the emotion vocabulary in the user data, including：
Emotion object vocabulary and emotion vocabulary are extracted in the user data based on default stacking CRFs models, had
Body includes：
The mark of part of speech is carried out to each vocabulary in user data using stacking CRFs models, including：
To the Chinese sentence X=(x provided1,x2,x3…xn), its part-of-speech tagging result sequence is found by viterbi algorithm
Y=(y1,y2,y3….yn), and make it that conditional probability P=(y | x) is maximum, and conditional probability P=(y | x) it is shown below：
Wherein：
Wherein, tkAnd slIt is characteristic function, λkAnd μlIt is corresponding weights, Z (x) is standardizing factor；
The emotion word in corpus of text and emotion object are carried out according to every words according to the part-of-speech rule of default participle
Statistics set.
Alternatively, it is described to be based on default benchmark emotion vocabulary, according to the emotion vocabulary of extraction and the benchmark emotion word
The similarity of remittance, the emotion vocabulary of the extraction is sorted out, including：
Vectorization is carried out to the benchmark emotion vocabulary included from language material, and clustered using K-means, is obtained each
Centre coordinate of the part of speech in default reference axis；
It is determined that coordinate of the emotion vocabulary extracted in default reference axis；
Calculate the Euclidean distance between the coordinate of the emotion vocabulary extracted and each centre coordinate；
If the coordinate and the first centre coordinate of the emotion vocabulary of the extraction are closest, the emotion vocabulary of the extraction
Belong to similar word with the benchmark emotion vocabulary corresponding to the first centre coordinate.
Alternatively, methods described also includes：
Multiple emotion object vocabulary and multiple emotion vocabulary to extraction carry out synonym classification processing.
Second aspect, the embodiment of the present invention provide a kind of individual feelings and emotions knowledge mapping and establish device again, including：
Acquiring unit, for obtaining the personal user data of knowledge mapping to be established, the user data includes and institute
State the text conversation and/or this paper questionnaires for being done of the individual of individual；
Extracting unit, for extracting emotion object vocabulary and emotion vocabulary, the emotion pair in the user data
The object limited as vocabulary by the emotion vocabulary；
Sort out unit, for based on default benchmark emotion vocabulary, according to the emotion vocabulary of extraction and the benchmark emotion
The similarity of vocabulary, the emotion vocabulary of the extraction is sorted out, and using the Sentiment orientation of generic vocabulary as described in
The Sentiment orientation of the emotion vocabulary of extraction；
Collection of illustrative plates generation unit, for building the emotion vocabulary of personal, the described emotion object vocabulary and the extraction
Sentiment orientation triple structure, using the triple structure as the personal individual feelings and emotions knowledge mapping.
Alternatively, the extracting unit, is further used for：
Emotion object vocabulary and emotion vocabulary are extracted in the user data based on default stacking CRFs models, had
Body includes：
The mark of part of speech is carried out to each vocabulary in user data using stacking CRFs models, including：
To the Chinese sentence X=(x provided1,x2,x3…xn), its part-of-speech tagging result sequence is found by viterbi algorithm
Y=(y1,y2,y3….yn), and make it that conditional probability P=(y | x) is maximum, and conditional probability P=(y | x) it is shown below：
Wherein：
Wherein, tkAnd slIt is characteristic function, λkAnd μlIt is corresponding weights, Z (x) is standardizing factor；
The emotion word in corpus of text and emotion object are carried out according to every words according to the part-of-speech rule of default participle
Statistics set.
Alternatively, the classification unit, is further used for：
Vectorization is carried out to the benchmark emotion vocabulary included from language material, and clustered using K-means, is obtained each
Centre coordinate of the part of speech in default reference axis；
It is determined that coordinate of the emotion vocabulary extracted in default reference axis；
Calculate the Euclidean distance between the coordinate of the emotion vocabulary extracted and each centre coordinate；
If the coordinate and the first centre coordinate of the emotion vocabulary of the extraction are closest, the emotion vocabulary of the extraction
Belong to similar word with the benchmark emotion vocabulary corresponding to the first centre coordinate.
Alternatively, the classification unit, is additionally operable to：
Multiple emotion object vocabulary and multiple emotion vocabulary to extraction carry out synonym classification processing.
The third aspect, another embodiment of the present invention provide a kind of computer equipment, including memory, processor and
Store the computer program that can be run on a memory and on a processor, it is characterised in that journey described in the computing device
The step of method as described in relation to the first aspect is realized during sequence.
Fourth aspect, another embodiment of the present invention provide a kind of computer-readable recording medium, are stored thereon with meter
Calculation machine program, it is characterised in that the step of method as described in relation to the first aspect is realized when the program is executed by processor.
The embodiments of the invention provide a kind of individual feelings and emotions knowledge mapping method for building up and device, computer equipment and meter
Calculation machine readable storage medium storing program for executing, in this method first from the personal text conversation and/or the personal this paper questionnaires done
Emotion object vocabulary and emotion vocabulary are extracted, then based on default benchmark emotion vocabulary, the emotion vocabulary of extraction is returned
Class, determine its Sentiment orientation, finally structure is personal, the triple structure of emotion object vocabulary and Sentiment orientation, and by this three
The tuple structure individual feelings and emotions knowledge mapping personal as this.So as to more complete according to the substantial amounts of language material structure of the individual
With for accurate emotion knowledge mapping, more valuable reference is provided for the application scenarios of each emotion knowledge mapping.
Brief description of the drawings
By reading the detailed description of hereafter preferred embodiment, it is various other the advantages of and benefit it is common for this area
Technical staff will be clear understanding.Accompanying drawing is only used for showing the purpose of preferred embodiment, and is not considered as to the present invention
Limitation.And in whole accompanying drawing, identical part is denoted by the same reference numerals.In the accompanying drawings：
Fig. 1 is a kind of individual feelings and emotions knowledge mapping method for building up embodiment flow chart provided by the invention；
Fig. 2 is that a kind of individual feelings and emotions knowledge mapping provided by the invention establishes device embodiment structural representation；
Fig. 3 is a kind of computer equipment example structure block diagram provided by the invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other under the premise of creative work is not made
Embodiment, belong to the scope of protection of the invention.
In a first aspect, the embodiments of the invention provide a kind of individual feelings and emotions knowledge mapping method for building up, as shown in figure 1, bag
Include：
S101, the personal user data for obtaining knowledge mapping to be established, the user data include with it is described personal
This paper questionnaires that text conversation and/or the individual are done；
S102, extraction emotion object vocabulary and emotion vocabulary, the emotion object vocabulary are in the user data
The object that the emotion vocabulary is limited；
S103, based on default benchmark emotion vocabulary, according to the phase of the emotion vocabulary of extraction and the benchmark emotion vocabulary
Like degree, the emotion vocabulary of the extraction is sorted out, and the feelings using the Sentiment orientation of generic vocabulary as the extraction
Feel the Sentiment orientation of vocabulary；
The Sentiment orientation of S104, structure personal, the described emotion object vocabulary and the extraction emotion vocabulary
Triple structure, using the triple structure as the personal individual feelings and emotions knowledge mapping.
The embodiments of the invention provide a kind of individual feelings and emotions knowledge mapping method for building up, first from the individual's in this method
Emotion object vocabulary and emotion vocabulary are extracted in text conversation and/or the personal this paper questionnaires done, then based on default
Benchmark emotion vocabulary, the emotion vocabulary of extraction is sorted out, determines its Sentiment orientation, finally structure is personal, emotion object
The triple structure of vocabulary and Sentiment orientation, and using the triple structure individual feelings and emotions knowledge mapping personal as this.From
And it can be chatted and be built more completely with being each emotion knowledge graph for accurate emotion knowledge mapping according to the substantial amounts of language of the individual
The application scenarios of spectrum provide more valuable reference.
Wherein, here can be chat record with the individual in some chat softwares with the text conversation of individual,
Or message in some social platforms etc..
Here emotion object vocabulary can be specific things, or the things of a certain type, for example, can
So that emotion object vocabulary is divided into several major classes, some specific objects can be included in every one kind.
1) emotion object vocabulary is object：Such as apple, green vegetables
2) emotion object vocabulary behaviour thing：Such as teacher, mother-in-law
3) emotion object vocabulary is event：For example work overtime, travel, reading
4) emotion object vocabulary is topic：Such as《Warwolf 2》Box office
In addition, default benchmark emotion vocabulary here can carry out adaptability setting according to being actually needed, can be several
Kind representative is used to state the vocabulary of mankind's class emotion, such as likes, and indignation is grieved, in terror, detests, anxious, nothing
Sense etc..The Sentiment orientation of such emotion vocabulary can it is identical from benchmark emotion vocabulary can also be different with benchmark emotion vocabulary.
In the specific implementation, can for extraction emotion object vocabulary and emotion vocabulary in user data in step S102
To implement in several ways, one of which optional embodiment is：
S1021, emotion object vocabulary and emotion extracted in the user data based on default stacking CRFs models
Vocabulary.
Specifically, Chinese sentence X=(x of the mark of part of speech exactly to providing is carried out using CRFs1,x2,x3…xn),
Its part-of-speech tagging result sequence Y=(y are found by viterbi algorithm1,y2,y3….yn) make conditional probability P=(y | x) maximum.
Wherein：
Wherein tkAnd slIt is characteristic function, λkAnd μlIt is corresponding weights, Z (x) is standardizing factor；
S1022, according to the part-of-speech rule of default participle by the emotion word in corpus of text and emotion object according to every
Words carry out statistics set.
Specifically：
Set A (can be the set of emotion main body), general only one.
The set B set of emotion object (can be), there are one or more synonyms in general a word.
The set C set of emotion word (can be), can be distributed in sentence with multiple, by the form of statistics by its
It is embodied in a set, does not require that its emotion tendency is identical.
In embodiments of the present invention, according to the actual requirements, the part-of-speech tagging of multilayer can be carried out to a word.First layer is first
The base attribute of word before this, such as adjective, verb, noun, adverbial word, preposition.The second layer marks grammer of the word in sentence
Composition, such as subject, predicate, object.Third layer represents.Third layer is the mark of emotion word and emotion subject word.
For example, table 1 shows that following sentence language material is segmented and marked multistage part of speech：
The language material of table 1 segments and marked multistage part of speech signal table
In the specific implementation, default benchmark emotion vocabulary is based in step S103, according to the emotion vocabulary of extraction and institute
The similarity of benchmark emotion vocabulary is stated, carrying out classification to the emotion vocabulary of the extraction can also implement in several ways,
The optional embodiment of one of which is：
S1031, vectorization is carried out to the benchmark emotion vocabulary included from language material, and clustered using K-means,
Obtain centre coordinate of each part of speech in default reference axis；
The coordinate of S1032, the emotion vocabulary for determining to extract in default reference axis；
Euclidean distance between S1033, the coordinate for calculating the emotion vocabulary extracted and each centre coordinate；
If S1034, the extraction emotion vocabulary coordinate and the first centre coordinate it is closest, the feelings of the extraction
Benchmark emotion vocabulary corresponding to sense vocabulary and the first centre coordinate belongs to similar word.
Specifically, all vocabulary in language material are collected and by its vectorization
Word according to including is clustered using K-means.Obtain the centre coordinate of each part of speech.
For near synonym, it can be carried out with the term vector after cluster Euclidean distance calculating and compared with, with its distance
Nearest is to represent its similar word.
Such as word X, Y.Vector (x is obtained after vectorization is carried out1,x2,x3…xn) and (y1,y2,y3….yn)。
Calculate the Euclidean distance formula between it：
For in general word, we calculate Euclidean distance in meaning of a word dictionary according to it to each word, take minimum to be similar
Word.
It is understandable to be, analyzed because method provided in an embodiment of the present invention can extract substantial amounts of language material, therefore
It is easy to be drawn into same or analogous emotion object vocabulary and corresponding emotion vocabulary in substantial amounts of expectation.Therefore this hair
The method that bright embodiment provides also includes：
Multiple emotion object vocabulary and multiple emotion vocabulary to extraction carry out synonym classification processing.
For example, include " tomato " and " tomato " in multiple emotion object vocabulary of extraction, it is evident that tomato
It is synonym with tomato, if be directed to the two words builds two triples respectively, then computing resource obviously can be wasted,
Therefore tomato and tomato can be merged into tomato herein.If judging, emotion vocabulary corresponding to the two falls within same emotion point
Class, namely corresponding Sentiment orientation are identical, then also merge the two corresponding emotion vocabulary, finally only form one this
Triple of the people for the hobby of tomato.
If judging, emotion vocabulary corresponding to the two belongs in different emotional semantic classifications, such as dialogue of the individual before one week
" I likes tomato " is mentioned, and " I dislikes tomato " is mentioned in the dialogue after one week.Tomato and tomato are carried out first
Merge, but " liking " clearly belongs to different emotional semantic classifications from " disagreeable ", at this moment can be by the posterior dialogue of time of origin
In emotion vocabulary corresponding to Sentiment orientation be considered Sentiment orientation of the individual for " tomato " this things, so as to make
The emotion knowledge mapping that must be generated more presses close to the current hobby state of the individual, to obtain more accurate knowledge mapping.
Second aspect, the embodiment of the present invention additionally provide a kind of individual feelings and emotions knowledge mapping and establish device, as shown in Fig. 2
Including：
Acquiring unit 201, for obtaining the personal user data of knowledge mapping to be established, the user data include with
This paper questionnaires that the personal text conversation and/or the individual are done；
Extracting unit 202, for extracting emotion object vocabulary and emotion vocabulary, the emotion in the user data
The object that object vocabulary is limited by the emotion vocabulary；
Sort out unit 203, for based on default benchmark emotion vocabulary, according to the emotion vocabulary of extraction and the benchmark feelings
Feel the similarity of vocabulary, the emotion vocabulary of the extraction is sorted out, and using the Sentiment orientation of generic vocabulary as institute
State the Sentiment orientation of the emotion vocabulary of extraction；
Collection of illustrative plates generation unit 204, for building the emotion word of personal, the described emotion object vocabulary and the extraction
The triple structure of the Sentiment orientation of remittance, using the triple structure as the personal individual feelings and emotions knowledge mapping.
Alternatively, the extracting unit 202, is further used for：
Emotion object vocabulary and emotion vocabulary are extracted in the user data based on default stacking CRFs models, had
Body includes：
The mark of part of speech is carried out to each vocabulary in user data using stacking CRFs models, including：
To the Chinese sentence X=(x provided1,x2,x3…xn), its part-of-speech tagging result sequence is found by viterbi algorithm
Y=(y1,y2,y3….yn), and make it that conditional probability P=(y | x) is maximum, and conditional probability P=(y | x) it is shown below：
Wherein：
Wherein, tkAnd slIt is characteristic function, λkAnd μlIt is corresponding weights, Z (x) is standardizing factor；
The emotion word in corpus of text and emotion object are carried out according to every words according to the part-of-speech rule of default participle
Statistics set.
Alternatively, the classification unit 203, is further used for：
Vectorization is carried out to the benchmark emotion vocabulary included from language material, and clustered using K-means, is obtained each
Centre coordinate of the part of speech in default reference axis；
It is determined that coordinate of the emotion vocabulary extracted in default reference axis；
Calculate the Euclidean distance between the coordinate of the emotion vocabulary extracted and each centre coordinate；
If the coordinate and the first centre coordinate of the emotion vocabulary of the extraction are closest, the emotion vocabulary of the extraction
Belong to similar word with the benchmark emotion vocabulary corresponding to the first centre coordinate.
Alternatively, the classification unit 203, is additionally operable to：
Multiple emotion object vocabulary and multiple emotion vocabulary to extraction carry out synonym classification processing.
By the individual feelings and emotions knowledge mapping that the present embodiment is introduced establishes device as that can perform in the embodiment of the present invention
Individual feelings and emotions knowledge mapping method for building up device, so based on the individual feelings and emotions knowledge graph described in the embodiment of the present invention
The method established is composed, the individual feelings and emotions knowledge mapping that those skilled in the art can understand the present embodiment establishes the tool of device
Body embodiment and its various change form, so establishing how device realizes this for the individual feelings and emotions knowledge mapping herein
Individual feelings and emotions knowledge mapping method for building up in inventive embodiments is no longer discussed in detail.As long as those skilled in the art implement
Device used by individual feelings and emotions knowledge mapping method for building up, belongs to the model to be protected of the application in the embodiment of the present invention
Enclose.
In addition, Fig. 3 shows the structured flowchart of computer equipment provided in an embodiment of the present invention.
Reference picture 3, the computer equipment, including：Processor (processor) 301, memory (memory) 302, with
And bus 303；
Wherein, the processor 301 and memory 302 complete mutual communication by the bus 303；
The processor 301 is used to call the programmed instruction in the memory 302, to perform above-mentioned each method embodiment
The method provided.
A kind of computer program product is also disclosed in the embodiment of the present invention, and the computer program product is non-temporary including being stored in
Computer program on state computer-readable recording medium, the computer program include programmed instruction, when described program instructs
When being computer-executed, computer is able to carry out the method that above-mentioned each method embodiment is provided.
The embodiment of the present invention also provides a kind of non-transient computer readable storage medium storing program for executing, and the non-transient computer is readable to deposit
Storage media stores computer instruction, and the computer instruction makes the computer perform the side that above-mentioned each method embodiment is provided
Method., such as including：
In the specification that this place provides, numerous specific details are set forth.It is to be appreciated, however, that the implementation of the present invention
Example can be put into practice in the case of these no details.In some instances, known method, structure is not been shown in detail
And technology, so as not to obscure the understanding of this description.
Similarly, it will be appreciated that in order to simplify the disclosure and help to understand one or more of each inventive aspect,
Above in the description to the exemplary embodiment of the present invention, each feature of the invention is grouped together into single implementation sometimes
In example, figure or descriptions thereof.However, the method for the disclosure should be construed to reflect following intention：I.e. required guarantor
The application claims of shield features more more than the feature being expressly recited in each claim.It is more precisely, such as following
Claims reflect as, inventive aspect is all features less than single embodiment disclosed above.Therefore,
Thus the claims for following embodiment are expressly incorporated in the embodiment, wherein each claim is in itself
Separate embodiments all as the present invention.
Those skilled in the art, which are appreciated that, to be carried out adaptively to the module in the equipment in embodiment
Change and they are arranged in one or more equipment different from the embodiment.Can be the module or list in embodiment
Member or component be combined into a module or unit or component, and can be divided into addition multiple submodule or subelement or
Sub-component.In addition at least some in such feature and/or process or unit exclude each other, it can use any
Combination is disclosed to all features disclosed in this specification (including adjoint claim, summary and accompanying drawing) and so to appoint
Where all processes or unit of method or equipment are combined.Unless expressly stated otherwise, this specification (including adjoint power
Profit requires, summary and accompanying drawing) disclosed in each feature can be by providing the alternative features of identical, equivalent or similar purpose come generation
Replace.
In addition, it will be appreciated by those of skill in the art that although some embodiments in this include institute in other embodiments
Including some features rather than further feature, but the combination of the feature of different embodiments means to be in the scope of the present invention
Within and form different embodiments.For example, in the following claims, embodiment claimed it is any it
One mode can use in any combination.
Some unit embodiments of the present invention can be realized with hardware, or to be run on one or more processor
Software module realize, or realized with combinations thereof.It will be understood by those of skill in the art that it can use in practice
Microprocessor or digital signal processor (DSP) realize gateway according to embodiments of the present invention, proxy server, system
In some or all parts some or all functions.The present invention is also implemented as described herein for performing
The some or all equipment or program of device (for example, computer program and computer program product) of method.So
Realization the present invention program can store on a computer-readable medium, or can have one or more signal shape
Formula.Such signal can be downloaded from internet website and obtained, and either be provided or with any other shape on carrier signal
Formula provides.
It should be noted that the present invention will be described rather than limits the invention for above-described embodiment, and ability
Field technique personnel can design alternative embodiment without departing from the scope of the appended claims.In the claims,
Any reference symbol between bracket should not be configured to limitations on claims.Word "comprising" does not exclude the presence of not
Element or step listed in the claims.Word "a" or "an" before element does not exclude the presence of multiple such
Element.The present invention can be by means of including the hardware of some different elements and being come by means of properly programmed computer real
It is existing.In if the unit claim of equipment for drying is listed, several in these devices can be by same hardware branch
To embody.The use of word first, second, and third does not indicate that any order.These words can be explained and run after fame
Claim.
Claims (10)
- A kind of 1. individual feelings and emotions knowledge mapping method for building up, it is characterised in that including：The personal user data of knowledge mapping to be established is obtained, the user data includes and the personal text conversation And/or this paper questionnaires that the individual is done；It is the emotion word that emotion object vocabulary and emotion vocabulary, the emotion object vocabulary are extracted in the user data The limited object of remittance；Based on default benchmark emotion vocabulary, according to the similarity of the emotion vocabulary of extraction and the benchmark emotion vocabulary, to institute The emotion vocabulary for stating extraction is sorted out, and the feelings using the Sentiment orientation of generic vocabulary as the emotion vocabulary of the extraction Sense tendency；Build the triple knot of the Sentiment orientation of personal, the described emotion object vocabulary and the emotion vocabulary of the extraction Structure, using the triple structure as the personal individual feelings and emotions knowledge mapping.
- 2. according to the method for claim 1, it is characterised in that described that emotion object vocabulary is extracted in the user data And emotion vocabulary, including：Emotion object vocabulary and emotion vocabulary are extracted in the user data based on default stacking CRFs models, specific bag Include：The mark of part of speech is carried out to each vocabulary in user data using stacking CRFs models, including：To the Chinese sentence X=(x provided1,x2,x3…xn), its part-of-speech tagging result sequence Y=is found by viterbi algorithm (y1,y2,y3….yn), and make it that conditional probability P=(y | x) is maximum, and conditional probability P=(y | x) it is shown below：<mrow> <mi>P</mi> <mrow> <mo>(</mo> <mi>y</mi> <mo>|</mo> <mi>x</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mn>1</mn> <mrow> <mi>Z</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>)</mo> </mrow> </mrow> </mfrac> <mi>exp</mi> <mrow> <mo>(</mo> <munder> <mo>&Sigma;</mo> <mrow> <mi>i</mi> <mo>,</mo> <mi>k</mi> </mrow> </munder> <msub> <mi>&lambda;</mi> <mi>k</mi> </msub> <msub> <mi>t</mi> <mi>k</mi> </msub> <mo>(</mo> <mrow> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>-</mo> <mn>1</mn> </mrow> </msub> <mo>,</mo> <msub> <mi>y</mi> <mi>i</mi> </msub> <mo>,</mo> <mi>x</mi> <mo>,</mo> <mi>i</mi> </mrow> <mo>)</mo> <mo>+</mo> <munder> <mo>&Sigma;</mo> <mrow> <mi>i</mi> <mo>,</mo> <mi>l</mi> </mrow> </munder> <msub> <mi>&mu;</mi> <mi>l</mi> </msub> <msub> <mi>s</mi> <mi>l</mi> </msub> <mo>(</mo> <mrow> <msub> <mi>y</mi> <mi>i</mi> </msub> <mo>,</mo> <mi>x</mi> <mo>,</mo> <mi>i</mi> </mrow> <mo>)</mo> <mo>)</mo> </mrow> </mrow>Wherein：<mrow> <mi>Z</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>)</mo> </mrow> <mo>=</mo> <munder> <mo>&Sigma;</mo> <mi>y</mi> </munder> <mi>exp</mi> <mrow> <mo>(</mo> <munder> <mo>&Sigma;</mo> <mrow> <mi>i</mi> <mo>,</mo> <mi>k</mi> </mrow> </munder> <msub> <mi>&lambda;</mi> <mi>k</mi> </msub> <msub> <mi>t</mi> <mi>t</mi> </msub> <mo>(</mo> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>-</mo> <mn>1</mn> </mrow> </msub> <mo>,</mo> <msub> <mi>y</mi> <mi>i</mi> </msub> <mo>,</mo> <mi>x</mi> <mo>,</mo> <mi>i</mi> <mo>)</mo> </mrow> <mo>+</mo> <munder> <mo>&Sigma;</mo> <mrow> <mi>i</mi> <mo>,</mo> <mi>l</mi> </mrow> </munder> <msub> <mi>&mu;</mi> <mi>l</mi> </msub> <msub> <mi>s</mi> <mi>l</mi> </msub> <mo>(</mo> <mrow> <msub> <mi>y</mi> <mi>i</mi> </msub> <mo>,</mo> <mi>x</mi> <mo>,</mo> <mi>i</mi> </mrow> <mo>)</mo> <mo>)</mo> </mrow>Wherein, tkAnd slIt is characteristic function, λkAnd μlIt is corresponding weights, Z (x) is standardizing factor；The emotion word in corpus of text and emotion object are counted according to every words according to the part-of-speech rule of default participle Set.
- 3. according to the method for claim 1, it is characterised in that it is described to be based on default benchmark emotion vocabulary, according to extraction Emotion vocabulary and the benchmark emotion vocabulary similarity, the emotion vocabulary of the extraction is sorted out, including：Vectorization is carried out to the benchmark emotion vocabulary included from language material, and clustered using K-means, obtains each part of speech Centre coordinate in default reference axis；It is determined that coordinate of the emotion vocabulary extracted in default reference axis；Calculate the Euclidean distance between the coordinate of the emotion vocabulary extracted and each centre coordinate；If the coordinate and the first centre coordinate of the emotion vocabulary of the extraction are closest, the emotion vocabulary of the extraction and Benchmark emotion vocabulary corresponding to one centre coordinate belongs to similar word.
- 4. according to the method for claim 1, it is characterised in that methods described also includes：Multiple emotion object vocabulary and multiple emotion vocabulary to extraction carry out synonym classification processing.
- 5. a kind of individual feelings and emotions knowledge mapping establishes device, it is characterised in that including：Acquiring unit, for obtaining the personal user data of knowledge mapping to be established, the user data includes and described This paper questionnaires that the text conversation of people and/or the individual are done；Extracting unit, for extracting emotion object vocabulary and emotion vocabulary, the emotion subject word in the user data Converge the object limited by the emotion vocabulary；Sort out unit, for based on default benchmark emotion vocabulary, according to the emotion vocabulary of extraction and the benchmark emotion vocabulary Similarity, the emotion vocabulary of the extraction is sorted out, and using the Sentiment orientation of generic vocabulary as the extraction Emotion vocabulary Sentiment orientation；Collection of illustrative plates generation unit, the feelings of the emotion vocabulary for building personal, the described emotion object vocabulary and the extraction The triple structure of tendency is felt, using the triple structure as the personal individual feelings and emotions knowledge mapping.
- 6. device according to claim 5, it is characterised in that the extracting unit, be further used for：Emotion object vocabulary and emotion vocabulary are extracted in the user data based on default stacking CRFs models, specific bag Include：The mark of part of speech is carried out to each vocabulary in user data using stacking CRFs models, including：To the Chinese sentence X=(x provided1,x2,x3…xn), its part-of-speech tagging result sequence Y=is found by viterbi algorithm (y1,y2,y3….yn), and make it that conditional probability P=(y | x) is maximum, and conditional probability P=(y | x) it is shown below：<mrow> <mi>P</mi> <mrow> <mo>(</mo> <mi>y</mi> <mo>|</mo> <mi>x</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mn>1</mn> <mrow> <mi>Z</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>)</mo> </mrow> </mrow> </mfrac> <mi>exp</mi> <mrow> <mo>(</mo> <munder> <mo>&Sigma;</mo> <mrow> <mi>i</mi> <mo>,</mo> <mi>k</mi> </mrow> </munder> <msub> <mi>&lambda;</mi> <mi>k</mi> </msub> <msub> <mi>t</mi> <mi>k</mi> </msub> <mo>(</mo> <mrow> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>-</mo> <mn>1</mn> </mrow> </msub> <mo>,</mo> <msub> <mi>y</mi> <mi>i</mi> </msub> <mo>,</mo> <mi>x</mi> <mo>,</mo> <mi>i</mi> </mrow> <mo>)</mo> <mo>+</mo> <munder> <mo>&Sigma;</mo> <mrow> <mi>i</mi> <mo>,</mo> <mi>l</mi> </mrow> </munder> <msub> <mi>&mu;</mi> <mi>l</mi> </msub> <msub> <mi>s</mi> <mi>l</mi> </msub> <mo>(</mo> <mrow> <msub> <mi>y</mi> <mi>i</mi> </msub> <mo>,</mo> <mi>x</mi> <mo>,</mo> <mi>i</mi> </mrow> <mo>)</mo> <mo>)</mo> </mrow> </mrow>Wherein：<mrow> <mi>Z</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>)</mo> </mrow> <mo>=</mo> <munder> <mo>&Sigma;</mo> <mi>y</mi> </munder> <mi>exp</mi> <mrow> <mo>(</mo> <munder> <mo>&Sigma;</mo> <mrow> <mi>i</mi> <mo>,</mo> <mi>k</mi> </mrow> </munder> <msub> <mi>&lambda;</mi> <mi>k</mi> </msub> <msub> <mi>t</mi> <mi>t</mi> </msub> <mo>(</mo> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>-</mo> <mn>1</mn> </mrow> </msub> <mo>,</mo> <msub> <mi>y</mi> <mi>i</mi> </msub> <mo>,</mo> <mi>x</mi> <mo>,</mo> <mi>i</mi> <mo>)</mo> </mrow> <mo>+</mo> <munder> <mo>&Sigma;</mo> <mrow> <mi>i</mi> <mo>,</mo> <mi>l</mi> </mrow> </munder> <msub> <mi>&mu;</mi> <mi>l</mi> </msub> <msub> <mi>s</mi> <mi>l</mi> </msub> <mo>(</mo> <mrow> <msub> <mi>y</mi> <mi>i</mi> </msub> <mo>,</mo> <mi>x</mi> <mo>,</mo> <mi>i</mi> </mrow> <mo>)</mo> <mo>)</mo> </mrow>Wherein, tkAnd slIt is characteristic function, λkAnd μlIt is corresponding weights, Z (x) is standardizing factor；The emotion word in corpus of text and emotion object are counted according to every words according to the part-of-speech rule of default participle Set.
- 7. device according to claim 5, it is characterised in that the classification unit, be further used for：Vectorization is carried out to the benchmark emotion vocabulary included from language material, and clustered using K-means, obtains each part of speech Centre coordinate in default reference axis；It is determined that coordinate of the emotion vocabulary extracted in default reference axis；Calculate the Euclidean distance between the coordinate of the emotion vocabulary extracted and each centre coordinate；If the coordinate and the first centre coordinate of the emotion vocabulary of the extraction are closest, the emotion vocabulary of the extraction and Benchmark emotion vocabulary corresponding to one centre coordinate belongs to similar word.
- 8. device according to claim 5, it is characterised in that the classification unit, be additionally operable to：Multiple emotion object vocabulary and multiple emotion vocabulary to extraction carry out synonym classification processing.
- 9. a kind of computer equipment, including memory, processor and storage can be run on a memory and on a processor Computer program, it is characterised in that such as Claims 1-4 any methods described is realized during the computing device described program The step of.
- 10. a kind of computer-readable recording medium, is stored thereon with computer program, it is characterised in that the program is by processor The step of methods described as any such as Claims 1-4 is realized during execution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711145598.3A CN107895027A (en) | 2017-11-17 | 2017-11-17 | Individual feelings and emotions knowledge mapping method for building up and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711145598.3A CN107895027A (en) | 2017-11-17 | 2017-11-17 | Individual feelings and emotions knowledge mapping method for building up and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107895027A true CN107895027A (en) | 2018-04-10 |
Family
ID=61805456
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711145598.3A Pending CN107895027A (en) | 2017-11-17 | 2017-11-17 | Individual feelings and emotions knowledge mapping method for building up and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107895027A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109065071A (en) * | 2018-08-31 | 2018-12-21 | 电子科技大学 | A kind of song clusters method based on Iterative k-means Algorithm |
| CN109840268A (en) * | 2018-12-23 | 2019-06-04 | 国网浙江省电力有限公司 | A kind of universe data map construction method based on enterprise information model |
| CN111506697A (en) * | 2019-01-30 | 2020-08-07 | 北京入思技术有限公司 | Cross-modal emotion knowledge graph construction method and device |
| CN111914556A (en) * | 2020-06-19 | 2020-11-10 | 合肥工业大学 | Emotion guiding method and system based on emotion semantic transfer map |
| CN112417177A (en) * | 2020-12-10 | 2021-02-26 | 清研灵智信息咨询(北京)有限公司 | Knowledge graph construction method based on investigation data |
| CN113965541A (en) * | 2018-12-25 | 2022-01-21 | 创新先进技术有限公司 | Conversation expression processing method and device |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060265090A1 (en) * | 2005-05-18 | 2006-11-23 | Kelly Conway | Method and software for training a customer service representative by analysis of a telephonic interaction between a customer and a contact center |
| US20080077569A1 (en) * | 2006-09-27 | 2008-03-27 | Yahoo! Inc., A Delaware Corporation | Integrated Search Service System and Method |
| CN102184232A (en) * | 2011-05-11 | 2011-09-14 | 北京航空航天大学 | Chinese vocabulary emotion modeling method based on pleasure, arousal and dominance (PAD) |
| CN103207855A (en) * | 2013-04-12 | 2013-07-17 | 广东工业大学 | Fine-grained sentiment analysis system and method specific to product comment information |
| CN103678564A (en) * | 2013-12-09 | 2014-03-26 | 国家计算机网络与信息安全管理中心 | Internet product research system based on data mining |
| CN105930368A (en) * | 2016-04-13 | 2016-09-07 | 深圳大学 | Emotion classification method and system |
| CN106649250A (en) * | 2015-10-29 | 2017-05-10 | 北京国双科技有限公司 | Method and device for identifying emotional new words |
-
2017
- 2017-11-17 CN CN201711145598.3A patent/CN107895027A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060265090A1 (en) * | 2005-05-18 | 2006-11-23 | Kelly Conway | Method and software for training a customer service representative by analysis of a telephonic interaction between a customer and a contact center |
| US20080077569A1 (en) * | 2006-09-27 | 2008-03-27 | Yahoo! Inc., A Delaware Corporation | Integrated Search Service System and Method |
| CN102184232A (en) * | 2011-05-11 | 2011-09-14 | 北京航空航天大学 | Chinese vocabulary emotion modeling method based on pleasure, arousal and dominance (PAD) |
| CN103207855A (en) * | 2013-04-12 | 2013-07-17 | 广东工业大学 | Fine-grained sentiment analysis system and method specific to product comment information |
| CN103678564A (en) * | 2013-12-09 | 2014-03-26 | 国家计算机网络与信息安全管理中心 | Internet product research system based on data mining |
| CN106649250A (en) * | 2015-10-29 | 2017-05-10 | 北京国双科技有限公司 | Method and device for identifying emotional new words |
| CN105930368A (en) * | 2016-04-13 | 2016-09-07 | 深圳大学 | Emotion classification method and system |
Non-Patent Citations (1)
| Title |
|---|
| 黄念娥: "本体与条件随机场结合的涉农商品名称抽取与类别标注", 《计算机应用》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109065071A (en) * | 2018-08-31 | 2018-12-21 | 电子科技大学 | A kind of song clusters method based on Iterative k-means Algorithm |
| CN109065071B (en) * | 2018-08-31 | 2021-05-14 | 电子科技大学 | Song clustering method based on iterative k-means algorithm |
| CN109840268A (en) * | 2018-12-23 | 2019-06-04 | 国网浙江省电力有限公司 | A kind of universe data map construction method based on enterprise information model |
| CN113965541A (en) * | 2018-12-25 | 2022-01-21 | 创新先进技术有限公司 | Conversation expression processing method and device |
| CN113965541B (en) * | 2018-12-25 | 2023-06-02 | 创新先进技术有限公司 | Conversation expression processing method and device |
| CN111506697A (en) * | 2019-01-30 | 2020-08-07 | 北京入思技术有限公司 | Cross-modal emotion knowledge graph construction method and device |
| CN111914556A (en) * | 2020-06-19 | 2020-11-10 | 合肥工业大学 | Emotion guiding method and system based on emotion semantic transfer map |
| CN111914556B (en) * | 2020-06-19 | 2023-10-31 | 合肥工业大学 | Emotion guiding method and system based on emotion semantic transfer pattern |
| CN112417177A (en) * | 2020-12-10 | 2021-02-26 | 清研灵智信息咨询(北京)有限公司 | Knowledge graph construction method based on investigation data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107895027A (en) | Individual feelings and emotions knowledge mapping method for building up and device | |
| CN108536681B (en) | Intelligent question-answering method, device, equipment and storage medium based on emotion analysis | |
| CN111767741B (en) | Text emotion analysis method based on deep learning and TFIDF algorithm | |
| US10949753B2 (en) | Causal modeling and attribution | |
| Apriyanto et al. | Born in social media culture: personality features impact in communication context | |
| CN108052505A (en) | Text emotion analysis method and device, storage medium, terminal | |
| US10496751B2 (en) | Avoiding sentiment model overfitting in a machine language model | |
| CN108845986A (en) | A kind of sentiment analysis method, equipment and system, computer readable storage medium | |
| Abdullah et al. | Sentiment analysis on arabic tweets: Challenges to dissecting the language | |
| CN110309513B (en) | Text dependency analysis method and device | |
| JP2013190985A (en) | Knowledge response system, method and computer program | |
| CN110324278A (en) | Account main body consistency detecting method, device and equipment | |
| CN104216880B (en) | Term based on internet defines discrimination method | |
| Shovon et al. | The performance of graph neural network in detecting fake news from social media feeds | |
| Dzieciątko | Application of text analytics to analyze emotions in the speeches | |
| KR20120117297A (en) | Method for summerizing meeting minutes based on sentence network | |
| Tho et al. | A comparison of lexicon-based and transformer-based sentiment analysis on code-mixed of low-resource languages | |
| Zhuo et al. | Chinese text sentiment analysis based on fuzzy semantic model | |
| JP6150664B2 (en) | Mining analyzer, method and program | |
| Huang et al. | You are what you tweet: A new hybrid model for sentiment analysis | |
| Bogović et al. | Topic Modeling for Tracking COVID-19 Communication on Twitter | |
| JP2021056857A (en) | Information processing system, information processing method, and program | |
| Tiwary et al. | Determination Of Category–Wise Influential Users Using Information Retrieval Technique from Twitter | |
| Cavalin et al. | Towards a Method to Classify Language Style for Enhancing Conversational Systems | |
| Wang et al. | Building sentimental word Lexicon for Chinese Movie comments |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180410 |