CN107943993A - A kind of method for learning Chinese and system based on complex network - Google Patents
A kind of method for learning Chinese and system based on complex network Download PDFInfo
- Publication number
- CN107943993A CN107943993A CN201711257335.1A CN201711257335A CN107943993A CN 107943993 A CN107943993 A CN 107943993A CN 201711257335 A CN201711257335 A CN 201711257335A CN 107943993 A CN107943993 A CN 107943993A
- Authority
- CN
- China
- Prior art keywords
- learning
- chinese character
- cost
- chinese
- word
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 230000008569 process Effects 0.000 claims abstract description 18
- 150000001875 compounds Chemical class 0.000 claims description 29
- 238000004364 calculation method Methods 0.000 claims description 19
- 230000008901 benefit Effects 0.000 claims description 14
- 238000004458 analytical method Methods 0.000 claims description 9
- 230000001186 cumulative effect Effects 0.000 claims description 9
- 238000010276 construction Methods 0.000 claims description 7
- 238000005194 fractionation Methods 0.000 claims description 7
- 238000013518 transcription Methods 0.000 claims description 7
- 230000035897 transcription Effects 0.000 claims description 7
- 240000004282 Grewia occidentalis Species 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 6
- 238000011160 research Methods 0.000 abstract description 6
- 230000000694 effects Effects 0.000 abstract description 3
- 239000000463 material Substances 0.000 abstract description 3
- 238000000265 homogenisation Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000000205 computational method Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000012797 qualification Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 235000010894 Artemisia argyi Nutrition 0.000 description 1
- 244000030166 artemisia Species 0.000 description 1
- 239000013065 commercial product Substances 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000003012 network analysis Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 239000002023 wood Substances 0.000 description 1
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
- G06F16/374—Thesaurus
-
- 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
- G06F16/34—Browsing; Visualisation therefor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/20—Education
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09B—EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS
- G09B19/00—Teaching not covered by other main groups of this subclass
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Educational Technology (AREA)
- Tourism & Hospitality (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Educational Administration (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Computational Linguistics (AREA)
- Document Processing Apparatus (AREA)
Abstract
The invention discloses a kind of method for learning Chinese and system based on complex network, this method is by analyzing the composition structure of Chinese character, design rational algorithm, build the component relationship network between Chinese character, and combine the complexity of Hanzi structure and use word frequency, the learning cost and income of Chinese character are calculated, provides the Chinese character learning order of science.In addition, China is a multi-ethnic country, except Chinese teaching, the succession of the spoken and written languages of national minorities is also an important component of modern teaching.Therefore, the present invention can also be generalized to during the teaching research of the spoken and written languages of national minorities.The present invention can be applied not only to the teaching process of spoken and written languages class, the selection of teaching material, in writing and evaluating, it can also be applied to business teaching software and the exploitation of APP so that language learner more scientific can reasonably be learnt, and obtain maximized learning effect.
Description
Technical field
The present invention relates to information industry and modern service industry technical field, particularly a kind of Chinese character study based on complex network
Learning method and system.
Background technology
The study and teaching of spoken and written languages are the work of one universality in the whole world, and the Words study for how formulating science is suitable
Sequence, it is the problem that teaching person pays close attention to for a long time learner is obtained maximum study income with minimum learning cost.
The rise of Network Science is the composition structure of our study texts, builds the relational network between words, calculates word
Lexicography practises costs and benefits, and the words learning sequence for formulating science provides new perspective in research and direction.Such as, Oxford,
R.L et al. establishes the English word learning order model of science by network analysis;Yan is small, and brave et al. to construct 3500 normal
With the network consisting of Chinese character, and give the learning sequence of these Chinese characters.
Both at home and abroad although the research to words teaching sequence has been achieved for some achievements, but still comes with some shortcomings it
Place.Words network is also confined on some common words, and the method for building network is also relied primarily on and traditional manually taken
Build, without scientific and reasonable computerized algorithm so that achievement in research is difficult to change into corresponding commercial product, it is more difficult to promotes
Into the learning process of other language.
The content of the invention
An object of the present invention is to propose a kind of method for learning Chinese based on complex network;The second object of the present invention
It is to propose a kind of Chinese character learning system based on complex network, while additionally provides Chinese character and disassemble and the meter of Chinese character learning cost
Calculation method.
An object of the present invention is achieved through the following technical solutions:A kind of Chinese character learning based on complex network
Method, comprises the following steps:
Establish Chinese character building block database:Chinese character building block includes phonetic transcriptions of Chinese characters and splits obtained phonetic composition portion
Part and Chinese character pattern split obtained font building block;
According to the composition structure of Chinese character in Chinese character building block database, word word relational network is constructed, each Chinese character is formed
A node in word word relational network, the component relationship between Chinese character form the side in word word relational network;
According to the word word relational network of construction, analyze the architectural feature of word word relational network and calculate its topological attribute value;
According to Chinese character building block database and word word relational network, the learning cost and study for calculating each Chinese character are received
Benefit;
According to position of the Chinese character in word word relational network, the costs and benefits of study, during obtaining actual learning
Sequencing, forms the learning roadmap of Chinese character;
The learning process and learning state and study route model of recording learning person;
According to the learning process and learning state of learner, the current Chinese character learning degree of analytic learning person, to follow-up
Chinese character learning route is updated.
Preferably, it is further comprising the steps of:
Export the current study schedule of learner and optimal Chinese character learning route.
Preferably, the phonetic building block is obtained according to following steps:
Chinese character is subjected to phonetic fractionation, it is specific as follows:
1) phonetic splits into initial consonant and simple or compound vowel of a Chinese syllable;
2) phonetic being individually made of simple or compound vowel of a Chinese syllable is not split.
Preferably, the font building block is obtained according to following steps:
1) radical is only used as, not independent into word, no pronunciation, without radical, does not split;
2) radical can be used as, and can be independently itself into word, radical, is not split;
3) it is radical and can be split with other Chinese combinatorics on words by radical;
4) to the not unique Chinese character of split result, compare phonetic and the four-corner system in the following way, take matching degree most
Big:
The matching algorithm of the matching degree is:If phonetic or corner is corresponding encodes identical, value 1, otherwise value
0。
Preferably, the learning cost of the Chinese character includes pinyin learning cost and font learning cost;
The pinyin learning cost specifically calculates in the following way:
According to phonetic split result, initial consonant, simple or compound vowel of a Chinese syllable respectively calculate 1 cost, polyphone cumulative calculation;
The font learning cost specifically calculates in the following way:
How much calculated according to stroke, cost is stroke number.
Preferably, saving cost calculation is further included, the saving cost calculation includes phonetic saving _ percent of total, phonetic
Saving _ subclass percent of total, phonetic saving _ subclass percentage, stroke saving _ percent of total, stroke saving _ subclass percentage,
Child node is maximum to save cost, father node always cost-effective calculating process;
Phonetic saving _ the percent of total, comprises the following steps that:
J is all nodes;
Wherein, if j only has with i, initial consonant or simple or compound vowel of a Chinese syllable are identical, and it is 1 to save cost;If initial consonant and simple or compound vowel of a Chinese syllable are all identical, save
Cost is 2, polyphone cumulative calculation;
Phonetic saving _ subclass the percent of total, comprises the following steps that:
J is the descendant nodes of i;
Phonetic saving _ subclass the percentage, comprises the following steps that:
J is the descendant nodes of i;
Stroke saving _ the percent of total, comprises the following steps that:
J is the descendant nodes of i;
Wherein, Chinese character j is the child node of i, and i is contained in font, so the stroke of j saves the stroke number that cost is i;
Stroke saving _ subclass the percentage, comprises the following steps that:
J is the descendant nodes of i;
The child node is maximum to save cost, comprises the following steps that:
max<The stroke of phonetic saving _ _ subclass percent of total+Chinese character j of Chinese character j saves percent of total>, j is that the sub of i is saved
Point;
The father node always saves cost, comprises the following steps that:
The stroke that the father node of Chinese character i always saves cost=∑ Chinese character k saves cost, and k is the father node of i.
The second object of the present invention is achieved through the following technical solutions, a kind of Chinese character learning based on complex network
System, including:
Chinese character building block database, is made of Chinese character building block, and Chinese character building block includes phonetic transcriptions of Chinese characters and splits institute
The phonetic building block and Chinese character pattern of acquisition split obtained font building block;
Word word relational network, forms structure, each Chinese character forms word word and close by Chinese character in Chinese character building block database
It is a node in network, the component relationship between Chinese character forms the side in word word relational network;
Nework analysis module:According to the word word relational network of construction, architectural feature and the calculating of word word relational network are analyzed
Its topological attribute value;
Cost-benefit module:According to Chinese character building block database and word word relational network, the study of each Chinese character is calculated
Cost and study income;
Learning roadmap:According to position of the Chinese character in word word relational network, the costs and benefits of study, actual is obtained
Sequencing during habit, forms the learning roadmap of Chinese character;
Historical data base:The learning process and learning state and study route model of recording learning person;
Study condition analysis module:According to the learning process and learning state of learner, the current Chinese character of analytic learning person
Level of learning, is updated follow-up Chinese character learning route.
Preferably, the phonetic building block is obtained according to following steps:
Chinese character is subjected to phonetic fractionation, it is specific as follows:
1) phonetic splits into initial consonant and simple or compound vowel of a Chinese syllable;
2) phonetic being individually made of simple or compound vowel of a Chinese syllable is not split.
Preferably, the font building block is obtained according to following steps:
1) radical is only used as, not independent into word, no pronunciation, without radical, does not split;
2) radical can be used as, and can be independently itself into word, radical, is not split;
3) it is radical and can be split with other Chinese combinatorics on words by radical;
4) to the not unique Chinese character of split result, compare phonetic and the four-corner system in the following way, take matching degree most
Big:
The matching algorithm of the matching degree is:If phonetic or corner is corresponding encodes identical, value 1, otherwise value
0。
Preferably, the learning cost of the Chinese character includes pinyin learning cost and font learning cost;
The pinyin learning cost specifically calculates in the following way:
According to phonetic split result, initial consonant, simple or compound vowel of a Chinese syllable respectively calculate 1 cost, polyphone cumulative calculation;
The font learning cost specifically calculates in the following way:
How much calculated according to stroke, cost is stroke number.
By adopting the above-described technical solution, the present invention has the advantage that:
The present invention designs rational algorithm, builds the group between Chinese character by analyzing the composition structure of Chinese character
Into relational network, and combine the complexity of Hanzi structure and using word frequency, calculate the learning cost and income of Chinese character, give
Go out the Chinese character learning order of science.In addition, China is a multi-ethnic country, except Chinese teaching, minority language text
The succession of word is also an important component of modern teaching.Therefore, the present invention can also be generalized to minority language text
During the teaching research of word.
The present invention can be applied not only to the teaching process of spoken and written languages class, and the selection of teaching material, in writing and evaluating, may be used also
With the exploitation applied to business teaching software and APP so that language learner more scientific can reasonably be learnt, and be taken
Obtain maximized learning effect.
Other advantages, target and the feature of the present invention will be illustrated in the following description to a certain extent, and
And to a certain extent, based on will be apparent to those skilled in the art to investigating hereafter, Huo Zheke
To be instructed from the practice of the present invention.The target and other advantages of the present invention can be realized by following specification
And acquisition.
Brief description of the drawings
The brief description of the drawings of the present invention is as follows.
Fig. 1 is the flow chart of the method for learning Chinese based on complex network of the present invention;
Fig. 2 is the Chinese character learning system functional block diagram based on complex network of the present invention;
Fig. 3 is the word word relational network figure in the embodiment of the present invention one.
Embodiment
The invention will be further described with reference to the accompanying drawings and examples.
Embodiment 1
As shown in the figure, the method for learning Chinese provided in this embodiment based on complex network, by the composition for analyzing Chinese character
Structure, constructs corresponding Chinese character relational network, using Complex Networks Analysis method, learning difficulty and use with reference to each word
Frequency, finally calculates the learning cost and study income of each word, so as to provide more effectively study route for Chinese character learning person
And learning method.
A kind of method for learning Chinese based on complex network, comprises the following steps:
S1 establishes Chinese character building block database:By way of machine adds manual intervention automatically, Chinese character is tied from composition
Split on structure, so that the Chinese character building block database after Chinese character separating is formed, for storing Chinese character building block and tearing open
Divide result.Chinese character building block includes the obtained phonetic building block of phonetic transcriptions of Chinese characters fractionation and Chinese character pattern splits and obtained
Font building block;
The Chinese character separating explanation of the present embodiment, it is specific as follows:
1. phonetic is split
1) phonetic splits into initial consonant and simple or compound vowel of a Chinese syllable, such as:Ba → b-a, bian → b-ian;
2) phonetic being individually made of simple or compound vowel of a Chinese syllable is not split, such as a → a;
2. font is split
1) radical is only used as, not independent into word, no pronunciation, without radical, does not split.Such as:Lv, Rui etc.;
2) radical can be used as, and can be independently into word, i.e. its radical is itself, is not split, such as:Mouthful, wood etc.;
3) radical and other Chinese combinatorics on words can be divided into, torn open by radical (according to stroke order being matched here)
Point.Such as:Ah → Fu+can, eh → mouth+Ah, etc..
It is not unique to split result, compare its phonetic and the four-corner system, take matching degree maximum.
Matching algorithm is:If phonetic or corner is corresponding encodes identical, value 1, otherwise value 0.
Such as:Pure white is removable to be divided into white+bare hill or white+,
Pure white, ai, 22617;
Bare hill, qi, 27717;
, qi, 22717;
Pure white is 2 with bare hill matching degree, and pure white is 3 with matching degree, thus the split result of pure white take it is white+;
3. it can not split or split result is unreasonable, it is necessary to be split by hand.It is mainly the following:
1) radical and another Chinese combinatorics on words can not be split as.Such as:Love, Ao, etc..
2) according to the 2nd article of method for splitting, can split successfully, but result is incorrect.Such as:Chinese mugwort be split as Lv+eight, Lv+people,
Lv+enter, but these three results are incorrect.
It 3) according to the 2nd article of method for splitting, can split into a variety of as a result, but not can determine that correct result in program.
Such as:Finish split result be than+fourth and ratio+ten,
Finish, bi, 22401;
Fourth, ding, 10200;
Ten, shi, 40000;
Bi Yuding, ten matching degree are 1, can not determine correct result.
S2 constructs word word relational network, each Chinese character structure according to the composition structure of Chinese character in Chinese character building block database
Into a node in word word relational network, the component relationship between Chinese character forms the side in word word relational network.Such as Fig. 3 institutes
Show:Mouthful and knife composition calls together, mouth, knife, call a node in respectively network together, and mouth and knife respectively have a line to be directed toward to call together.
S3 analyzes the architectural feature of word word relational network and calculates its topological attribute according to the word word relational network of construction
Value.Specifically, K-core decomposition is carried out to the Chinese character complex network, so that the hierarchical structure produced.
S4 according to Chinese character building block database and word word relational network, receive by the learning cost and study for calculating each Chinese character
Benefit.
Learning cost provided in this embodiment calculates, specific as follows:
1. pinyin learning cost.According to phonetic split result, initial consonant, simple or compound vowel of a Chinese syllable respectively calculate 1 cost, polyphone cumulative calculation.
Such as, phonetic a costs are 1;Eight, phonetic ba cost are 2;Just, phonetic bian and pian costs are 4.
2. font learning cost.How much calculated according to stroke, cost is stroke number.
Such as, Ah totally 7 draws, cost 7.
The saving cost calculation that the present embodiment also provides, it is specific as follows:
It is to learn to receive the saving value of the learning cost of overall or specific Chinese character, i.e. study after a Chinese character to save cost
Benefit.The network structure formed to be saved after cost is mainly split according to font to be calculated, phonetic and font calculate respectively, and
Carry out homogenization processing.Computational methods are as follows:
1. phonetic saving _ percent of total
The value is without considering network structure, influence of one phonetic of association to other phonetic transcriptions of Chinese characters.Calculation formula:
J is all nodes;
Explanation:If j only has with i, initial consonant or simple or compound vowel of a Chinese syllable are identical, and it is 1 to save cost;If initial consonant and simple or compound vowel of a Chinese syllable are all identical, save
Cost is 2, polyphone cumulative calculation.
2. phonetic saving _ subclass percent of total
Learn influence of the phonetic to the phonetic of its child node.Calculation formula:
J is the descendant nodes of i;
3. phonetic saving _ subclass percentage
J is the descendant nodes of i;
4. stroke saving _ percent of total
J is the descendant nodes of i;
Explanation:Chinese character j is the child node of i, and i is contained in font, so the stroke of j saves the stroke number that cost is i.
5. stroke saving _ subclass percentage
J is the descendant nodes of i;
6. child node is maximum to save cost
The maximum saving cost of the child node of Chinese character i=
max<The stroke of phonetic saving _ _ subclass percent of total+Chinese character j of Chinese character j saves percent of total>, j is the child node of i;
7. father node always saves cost
The stroke that the father node of Chinese character i always saves cost=∑ Chinese character k saves cost, and k is the father node of i;
Specifically, the study income of a Chinese character namely learns the saving value to global learning cost after the Chinese character, bag
Include two parts:
1. the calculating of phonetic saving value
Cpi:The pinyin learning cost of i
Cp:The total pinyin learning cost of all Chinese characters, Cp=∑iCpi
Spij:After learning i, the pinyin learning cost variable quantity of j, if i and j initial consonants, simple or compound vowel of a Chinese syllable are different, Spij=0;If i
There is identical a, S with j initial consonants or simple or compound vowel of a Chinese syllablepij=1;If i with j initial consonants are identical with simple or compound vowel of a Chinese syllable, Spij=2
Lij:If i is the part of j, i.e., in Chinese character network, there are the path that i reaches j, then Lij=1, otherwise Lij=
0
Spi:Cost is saved to the phonetic of overall Chinese character after study i after homogenization
2. similarly, the calculating of font saving value
Czi:The font learning cost of i
Cz:The total font learning cost of all Chinese characters, Cz=∑iCzi
Szij:After learning i, cost variable quantities of the j on font, if j includes i, Szij=Czi, otherwise Szij=0
Lij:If i is the part of j, i.e., in Chinese character network, there are the path that i reaches j, then Lij=1, otherwise Lij=
0
Szi:Saving costs of the Chinese character i on font after homogenization
With reference to (1) and (2), it can be deduced that total saving value at cost of Chinese character i is:
Example illustrated above illustrates.
According to the method for cost accounting, the phonetic cost and font cost such as following table of each Chinese character in figure:
Mouthful | Knife | Call together | Day | It is clear | Xiangxi | According to | It is total | |
Phonetic cost Cpi | 2 | 2 | 4 | 2 | 2 | 0 | 2 | Cp=14 |
Font cost Czi | 3 | 2 | 5 | 4 | 9 | 4 | 13 | Cz=30 |
Chinese character learning route map:According to each Chinese character status in a network, the costs and benefits of study, actual is obtained
Sequencing during habit;By taking " knife " as an example, after study " knife ", the variable quantity and learning sequence of other Chinese character learning costs:
Mouthful | Call together | Day | It is clear | Xiangxi | According to | |
Phonetic reduces cost | 0 | 2 | 0 | 1 | 0 | 1 |
Font reduces cost | 0 | 3 | 0 | 3 | 0 | 3 |
Explanation:
" knife " and " there is no component relationship between mouth, day, Xiangxi ", therefore think that " mouth, day, Xiangxi " do not have shadow to study knife to study
Ring;
The simple or compound vowel of a Chinese syllable of " knife " is identical with " clear, according to ", therefore the learning cost 1 of " clear, according to " is reduced on phonetic;
" calling together " is polyphone, and the simple or compound vowel of a Chinese syllable of two pronunciations is all identical with " knife ", therefore " knife " can be reduced " calling together " on phonetic
Cost be 2.
According to formula (1) and (2), the saving value of " knife " to overall Chinese character on phonetic and font is:
Similarly, saving value of each Chinese character to global learning cost in figure can be calculated
Phonetic saves total value | Font saves total value | It is total | |
Mouthful | 0 | 3/30+3/30+3/30=1/10 | 1/10=0.1 |
Knife | 2/7 | 2/30+2/30+2/30=1/15 | 37/105=0.352 |
Call together | 2/7 | 5/30+5/30=1/5 | 17/35=0.486 |
Day | 0 | 4/30+4/30=2/15 | 2/15=0.133 |
It is clear | 1/7 | 9/30=3/10 | 31/70=0.443 |
Xiangxi | 0 | 4/30=2/15 | 2/15=0.133 |
According to | 0 | 0 | 0 |
" photograph " is located at network orlop, sin prole superstite node, it is believed that study " photograph " the cost for not learning other Chinese characters
Have an impact, preference learning will not bring the reduction of holistic cost.
Thus, we can show that the preliminary of all Chinese characters is ordered as in the network:
Sequence | 1 | 2 | 3 | 4 | 5 | 6 | 7 |
Chinese character | Call together | It is clear | Knife | Day | Xiangxi | Mouthful | According to |
But it is many similar to word as " photograph " in overall Chinese character, in order to provide the learning sequence of this kind of Chinese character, I
Using the sum of cost savings value of its class node (class node of " photograph " is " clear " and " Xiangxi ") be its carry out two minor sorts.
In this example, when carrying out two minor sorts to " photograph ", the saving value that we define " photograph " is worth it for the saving of " clear " and " Xiangxi "
With i.e. 31/70+2/15=121/210=0.576.Here, we it is believed that:If the building block of a Chinese character
Finish, then learning the cost of the Chinese character will change to minimum, and preference learning will not cause holistic cost to increase.
In addition, the result of calculation in upper table, it can be seen that the cost savings value highest of " calling together ", and it is used as its composition
" knife " and " mouth " of component, but after " calling together ", this is obviously improper for sequence, can be obvious because preference learning " knife " and " mouth "
The learning cost of " calling together " is reduced, so as to reduce global learning cost.Therefore, the row calculated for Chinese character saving cost algorithm
Sequence is as a result, we should give appropriate and reasonably adjust.It is believed that the building block of Chinese character should prior to the Chinese character into
Row study.Therefore, when being ranked up, the building block of Chinese character should be adjusted to before Chinese character by we.By adjusting, with
The ranking results of upper Chinese character network are:
Sequence | 1 | 2 | 3 | 4 | 5 | 6 | 7 |
Chinese character | Knife | Mouthful | Call together | Day | It is clear | Xiangxi | According to |
S5 is according to position of the Chinese character in word word relational network, the costs and benefits of study, during obtaining actual learning
Sequencing, form the learning roadmap of Chinese character.
The learning process and learning state and study route model of S6 recording learning persons.
S7 is according to the learning process and learning state of learner, the current Chinese character learning degree of analytic learning person, to follow-up
Chinese character learning route be updated.Namely the data of Chinese character and qualification are grasped in Chinese character learning according to Chinese character learning person
The study route of optimization is found in route map;According to above ordering rule, we grasp feelings by testing the Chinese character of learner
Condition come recommend learn route.Such as learner " clear " word is not grasped, system can recommend " call together, day, clear " in order to help to learn
Habit person completes the study to " clear " word.
S8 exports the current study schedule of learner and optimal Chinese character learning route.
The present invention designs rational algorithm, builds the group between Chinese character by analyzing the composition structure of Chinese character
Into relational network, and combine the complexity of Hanzi structure and using word frequency, calculate the learning cost and income of Chinese character, give
Go out the Chinese character learning order of science.In addition, China is a multi-ethnic country, except Chinese teaching, minority language text
The succession of word is also an important component of modern teaching.Therefore, the present invention can also be generalized to minority language text
During the teaching research of word.
The present invention can be applied not only to the teaching process of spoken and written languages class, and the selection of teaching material, in writing and evaluating, may be used also
With the exploitation applied to business teaching software and APP so that language learner more scientific can reasonably be learnt, and be taken
Obtain maximized learning effect.
Embodiment two
The present invention also provides a kind of Chinese character learning system based on complex network, including:
Chinese character building block database, is made of Chinese character building block, and Chinese character building block includes phonetic transcriptions of Chinese characters and splits institute
The phonetic building block and Chinese character pattern of acquisition split obtained font building block.
Specifically, the phonetic building block is obtained according to following steps:
Chinese character is subjected to phonetic fractionation, it is specific as follows:
1) phonetic splits into initial consonant and simple or compound vowel of a Chinese syllable;
2) phonetic being individually made of simple or compound vowel of a Chinese syllable is not split.
The font building block is obtained according to following steps:
1) radical is only used as, not independent into word, no pronunciation, without radical, does not split;
2) radical can be used as, and can be independently itself into word, radical, is not split;
3) it is radical and can be split with other Chinese combinatorics on words by radical;
4) to the not unique Chinese character of split result, compare phonetic and the four-corner system in the following way, take matching degree most
Big:
The matching algorithm of the matching degree is:If phonetic or corner is corresponding encodes identical, value 1, otherwise value
0。
Word word relational network, forms structure, each Chinese character forms word word and close by Chinese character in Chinese character building block database
It is a node in network, the component relationship between Chinese character forms the side in word word relational network;
Nework analysis module:According to the word word relational network of construction, architectural feature and the calculating of word word relational network are analyzed
Its topological attribute value;
Cost-benefit module:According to Chinese character building block database and word word relational network, the study of each Chinese character is calculated
Cost and study income.Specifically, the learning cost of the Chinese character includes pinyin learning cost and font learning cost;
The pinyin learning cost specifically calculates in the following way:
According to phonetic split result, initial consonant, simple or compound vowel of a Chinese syllable respectively calculate 1 cost, polyphone cumulative calculation;
The font learning cost specifically calculates in the following way:
How much calculated according to stroke, cost is stroke number.
Learning roadmap:According to position of the Chinese character in word word relational network, the costs and benefits of study, actual is obtained
Sequencing during habit, forms the learning roadmap of Chinese character;
Historical data base:The learning process and learning state and study route model of recording learning person;Learner
The study route calculation function that route model is learner's historical data and the study condition analysis based on studying history storehouse is practised,
The model construction of study route is carried out to Limited Chinese Character Set.
Study condition analysis module:According to the learning process and learning state of learner, the current Chinese character of analytic learning person
Level of learning, is updated follow-up Chinese character learning route.Specifically, the phase of Chinese character learning person's current learning states is calculated
Data parameters are closed, the optimal study route of learner are calculated accordingly, so as to form the teaching plan of different learners;Study
The analysis of person's study condition relies on study router algorithm work, therefore the calculating of optimal study route refer to above institute
State the computational methods of learning roadmap.
Optimal Chinese character learning route recommendation system:For grasping the data of Chinese character and qualification according to Chinese character learning person
The study route of optimization is found in Chinese character learning route map;According to above ordering rule, we are by testing learner's
Chinese character grasps situation to recommend to learn route.Such as learner " clear " word is not grasped, system can recommend in order " call together, day,
It is clear " help the learner to complete study to " clear " word.
Finally illustrate, the above embodiments are merely illustrative of the technical solutions of the present invention and it is unrestricted, although with reference to compared with
The present invention is described in detail in good embodiment, it will be understood by those of ordinary skill in the art that, can be to the skill of the present invention
Art scheme technical scheme is modified or replaced equivalently, without departing from the objective and scope of the technical program, it should all cover in the present invention
Protection domain among.
Claims (10)
- A kind of 1. method for learning Chinese based on complex network, it is characterised in that:Comprise the following steps:Establish Chinese character building block database:Chinese character building block include phonetic transcriptions of Chinese characters split obtained phonetic building block and Chinese character pattern splits obtained font building block;According to the composition structure of Chinese character in Chinese character building block database, word word relational network is constructed, each Chinese character forms word word A node in relational network, the component relationship between Chinese character form the side in word word relational network;According to the word word relational network of construction, analyze the architectural feature of word word relational network and calculate its topological attribute value;According to Chinese character building block database and word word relational network, calculate the learning cost of each Chinese character and learn income;According to position of the Chinese character in word word relational network, the costs and benefits of study, the priority during actual learning is obtained Sequentially, the learning roadmap of Chinese character is formed;The learning process and learning state and study route model of recording learning person;According to the learning process and learning state of learner, the current Chinese character learning degree of analytic learning person, to follow-up Chinese character Study route is updated.
- A kind of 2. method for learning Chinese based on complex network according to claim 1, it is characterised in that:Further include following Step:Export the current study schedule of learner and optimal Chinese character learning route.
- A kind of 3. method for learning Chinese based on complex network according to claim 1, it is characterised in that:The pinyin-group Obtained into component according to following steps:Chinese character is subjected to phonetic fractionation, it is specific as follows:1) phonetic splits into initial consonant and simple or compound vowel of a Chinese syllable;2) phonetic being individually made of simple or compound vowel of a Chinese syllable is not split.
- A kind of 4. method for learning Chinese based on complex network according to claim 1, it is characterised in that:The glyph group Obtained into component according to following steps:1) radical is only used as, not independent into word, no pronunciation, without radical, does not split;2) radical can be used as, and can be independently itself into word, radical, is not split;3) it is radical and can be split with other Chinese combinatorics on words by radical;4) to the not unique Chinese character of split result, compare phonetic and the four-corner system in the following way, take matching degree maximum:The matching algorithm of the matching degree is:If phonetic or corner is corresponding encodes identical, value 1, otherwise value 0.
- A kind of 5. method for learning Chinese based on complex network according to claim 1, it is characterised in that:The Chinese character Learning cost includes pinyin learning cost and font learning cost;The pinyin learning cost specifically calculates in the following way:According to phonetic split result, initial consonant, simple or compound vowel of a Chinese syllable respectively calculate 1 cost, polyphone cumulative calculation;The font learning cost specifically calculates in the following way:How much calculated according to stroke, cost is stroke number.
- A kind of 6. method for learning Chinese based on complex network according to claim 1, it is characterised in that:Further include saving Cost calculation, the saving cost calculation include phonetic saving _ percent of total, phonetic saving _ subclass percent of total, phonetic section Province _ subclass percentage, stroke saving _ percent of total, stroke saving _ subclass percentage, the maximum saving cost of child node, father node Total cost-effective calculating process;Phonetic saving _ the percent of total, comprises the following steps that:J is all nodes;Wherein, if j only has with i, initial consonant or simple or compound vowel of a Chinese syllable are identical, and it is 1 to save cost;If initial consonant and simple or compound vowel of a Chinese syllable are all identical, cost is saved For 2, polyphone cumulative calculation;Phonetic saving _ subclass the percent of total, comprises the following steps that:J is the descendant nodes of i;Phonetic saving _ subclass the percentage, comprises the following steps that:J is the descendant nodes of i;Stroke saving _ the percent of total, comprises the following steps that:J is the descendant nodes of i;Wherein, Chinese character j is the child node of i, and i is contained in font, so the stroke of j saves the stroke number that cost is i;Stroke saving _ subclass the percentage, comprises the following steps that:J is the descendant nodes of i;The child node is maximum to save cost, comprises the following steps that:The maximum saving cost of the child node of Chinese character i=max<The stroke of the phonetic saving of Chinese character j _ subclass percent of total+Chinese character j saves percent of total>, j is the child node of i;The father node always saves cost, comprises the following steps that:The stroke that the father node of Chinese character i always saves cost=∑ Chinese character k saves cost, and k is the father node of i.
- A kind of 7. Chinese character learning system based on complex network, it is characterised in that:IncludingChinese character building block database, is made of Chinese character building block, and Chinese character building block includes phonetic transcriptions of Chinese characters fractionation and obtained Phonetic building block and Chinese character pattern split obtained font building block;Word word relational network, forms structure, each Chinese character forms word word network of personal connections by Chinese character in Chinese character building block database A node in network, the component relationship between Chinese character form the side in word word relational network;Nework analysis module:According to the word word relational network of construction, analyze the architectural feature of word word relational network and calculate it and open up Flutter property value;Cost-benefit module:According to Chinese character building block database and word word relational network, the study of each Chinese character is calculated Cost and study income;Learning roadmap:According to position of the Chinese character in word word relational network, the costs and benefits of study, actual learning mistake is obtained Sequencing in journey, forms the learning roadmap of Chinese character;Historical data base:The learning process and learning state and study route model of recording learning person;Study condition analysis module:According to the learning process and learning state of learner, the current Chinese character learning of analytic learning person Degree, is updated follow-up Chinese character learning route.
- A kind of 8. method for learning Chinese based on complex network according to claim 7, it is characterised in that:The pinyin-group Obtained into component according to following steps:Chinese character is subjected to phonetic fractionation, it is specific as follows:1) phonetic splits into initial consonant and simple or compound vowel of a Chinese syllable;2) phonetic being individually made of simple or compound vowel of a Chinese syllable is not split.
- A kind of 9. method for learning Chinese based on complex network according to claim 7, it is characterised in that:The glyph group Obtained into component according to following steps:1) radical is only used as, not independent into word, no pronunciation, without radical, does not split;2) radical can be used as, and can be independently itself into word, radical, is not split;3) it is radical and can be split with other Chinese combinatorics on words by radical;4) to the not unique Chinese character of split result, compare phonetic and the four-corner system in the following way, take matching degree maximum:The matching algorithm of the matching degree is:If phonetic or corner is corresponding encodes identical, value 1, otherwise value 0.
- A kind of 10. method for learning Chinese based on complex network according to claim 7, it is characterised in that:The Chinese character Learning cost include pinyin learning cost and font learning cost;The pinyin learning cost specifically calculates in the following way:According to phonetic split result, initial consonant, simple or compound vowel of a Chinese syllable respectively calculate 1 cost, polyphone cumulative calculation;The font learning cost specifically calculates in the following way:How much calculated according to stroke, cost is stroke number.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711257335.1A CN107943993B (en) | 2017-12-04 | 2017-12-04 | Chinese character learning method and system based on complex network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711257335.1A CN107943993B (en) | 2017-12-04 | 2017-12-04 | Chinese character learning method and system based on complex network |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107943993A true CN107943993A (en) | 2018-04-20 |
CN107943993B CN107943993B (en) | 2021-12-28 |
Family
ID=61948472
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711257335.1A Active CN107943993B (en) | 2017-12-04 | 2017-12-04 | Chinese character learning method and system based on complex network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107943993B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111508289A (en) * | 2020-04-14 | 2020-08-07 | 上海句石智能科技有限公司 | Language learning system based on word use frequency |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2000187657A (en) * | 1998-12-21 | 2000-07-04 | Fujitsu Fip Corp | Document editing device and recording medium recording document editing processing program |
CN1645356A (en) * | 2004-06-14 | 2005-07-27 | 曲万波 | Multiple dimensional Chinese studying systems |
KR101061420B1 (en) * | 2009-06-10 | 2011-09-01 | (주)칭화교육원 | How to Learn a Foreign Language Using the Internet |
CN106354701A (en) * | 2016-08-30 | 2017-01-25 | 腾讯科技(深圳)有限公司 | Chinese character processing method and device |
-
2017
- 2017-12-04 CN CN201711257335.1A patent/CN107943993B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2000187657A (en) * | 1998-12-21 | 2000-07-04 | Fujitsu Fip Corp | Document editing device and recording medium recording document editing processing program |
CN1645356A (en) * | 2004-06-14 | 2005-07-27 | 曲万波 | Multiple dimensional Chinese studying systems |
KR101061420B1 (en) * | 2009-06-10 | 2011-09-01 | (주)칭화교육원 | How to Learn a Foreign Language Using the Internet |
CN106354701A (en) * | 2016-08-30 | 2017-01-25 | 腾讯科技(深圳)有限公司 | Chinese character processing method and device |
Non-Patent Citations (3)
Title |
---|
DEHUN ZHAO、KUN ZHAO: ""A Complex Network Model Based on "Friends of a"", 《2014 IEEE 17TH INTERNATIONAL CONFERENCE ON COMPUTATIONAL SCIENCE AND ENGINEERING》 * |
JIANYU LI,JIE ZHOU: ""Chinese character structure analysis based on complex networks"", 《道客巴巴-HTTPS://WWW.DOC88.COM/P-3367767573542.HTML》 * |
王建伟、荣莉莉: ""基于复杂网络理论的中文字字网络的实证研究"", 《大连海事大学学报》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111508289A (en) * | 2020-04-14 | 2020-08-07 | 上海句石智能科技有限公司 | Language learning system based on word use frequency |
CN111508289B (en) * | 2020-04-14 | 2021-10-08 | 上海句石智能科技有限公司 | Language learning system based on word use frequency |
Also Published As
Publication number | Publication date |
---|---|
CN107943993B (en) | 2021-12-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108763326B (en) | Emotion analysis model construction method of convolutional neural network based on feature diversification | |
CN107291783B (en) | Semantic matching method and intelligent equipment | |
CN107818164A (en) | A kind of intelligent answer method and its system | |
CN104750672B (en) | A kind of Chinese vocabulary error correction method and its device being applied in search | |
CN108984530A (en) | A kind of detection method and detection system of network sensitive content | |
CN107644062B (en) | Knowledge content weight analysis system and method based on knowledge graph | |
CN105677822A (en) | Enrollment automatic question-answering method and system based on conversation robot | |
CN107862087A (en) | Sentiment analysis method, apparatus and storage medium based on big data and deep learning | |
CN107943784A (en) | Relation extraction method based on generation confrontation network | |
CN110825850B (en) | Natural language theme classification method and device | |
CN108108347B (en) | Dialogue mode analysis system and method | |
CN110096587A (en) | The fine granularity sentiment classification model of LSTM-CNN word insertion based on attention mechanism | |
CN104778160B (en) | A kind of english composition content is kept to the point analysis method | |
CN106909573A (en) | A kind of method and apparatus for evaluating question and answer to quality | |
CN112132536A (en) | Post recommendation method, system, computer equipment and storage medium | |
CN107066452A (en) | Translate householder method, translation servicing unit, translating equipment and translation auxiliary program | |
CN107943993A (en) | A kind of method for learning Chinese and system based on complex network | |
CN107894976A (en) | A kind of mixing language material segmenting method based on Bi LSTM | |
CN107797986A (en) | A kind of mixing language material segmenting method based on LSTM CNN | |
CN110147550A (en) | Pronunciation character fusion method neural network based | |
CN109284373A (en) | The acquisition methods and device of product up-gradation strategy based on text mining driving | |
CN106126497A (en) | A kind of automatic mining correspondence executes leader section and the method for cited literature textual content fragment | |
CN106446264B (en) | Document representation method and system | |
CN106909536A (en) | Method is recommended in a kind of scoring based on Heterogeneous Information | |
CN109408637A (en) | A kind of method and system automatically parsing list of engineering |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |