CN104281690A - Tag cloud generating method and device - Google Patents
Tag cloud generating method and device Download PDFInfo
- Publication number
- CN104281690A CN104281690A CN201410534723.XA CN201410534723A CN104281690A CN 104281690 A CN104281690 A CN 104281690A CN 201410534723 A CN201410534723 A CN 201410534723A CN 104281690 A CN104281690 A CN 104281690A
- Authority
- CN
- China
- Prior art keywords
- label
- article
- matrix
- cloud
- section
- 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
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
- G06F16/31—Indexing; Data structures therefor; Storage structures
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a tag cloud generating method and device. The tag cloud generating method includes steps of generating requests by receiving tag clouds with article set information, calculating weights of the tags in their belonging articles to generate an article-tag matrix according to the various tags in each article in the article set corresponding to the article set information, subjecting the article-tag matrix to singular value decomposition and generating a first matrix indicating the weights of the feature vectors in the article set and a second matrix indicating the weight of the tags in the feature vectors; generating the tag clouds according to the first matrix, the second matrix and preset generation rules. In the tag cloud generating method, the article-tag matrix is subjected to singular value decomposition and the tag clouds are generated via the decomposed matrixes, so that the problems that when the tag clouds generated by the prior art are used as indexes of key contents of the article set, the semantic range indicated by each tag element covers too much and the key contents of the article set are inaccurate are solved.
Description
Technical field
The application relates to label-cloud technical field, particularly relates to a kind of label-cloud generation method and device.
Background technology
Label-cloud usually by some article set of index, and then utilizes the label that in this article set, the frequency of occurrences is higher to generate.By this label-cloud visual means is shown, user can be allowed intuitively to understand information important in this article set, and this label-cloud also can be used as the index of article set key content, when user clicks any one tag element in this label-cloud, the information relevant to this tag element can be found out immediately from article set, facilitate user to consult information.
But; because traditional label-cloud carries out on the basis of frequency statistics direct to each label in article set; the each label utilizing the frequency to meet preset requirement generates as the tag element of in label-cloud; therefore; during the index that usual meeting causes label-cloud as article set key content because of the tag element form single (that is: each tag element is only made up of a label) of composition label-cloud, the inadequate problem accurately of the key content that is too wide in range, that embody article set of the semantic coverage indicated by each tag element.
Summary of the invention
In view of this, the application provides a kind of label-cloud generation method and device, the key content problem not accurately that when label-cloud generated to avoid prior art is as the index of article set key content, the semantic coverage indicated by each tag element is too wide in range, embody article set.
To achieve these goals, the technical scheme that provides of the embodiment of the present invention is as follows:
A kind of label-cloud generation method, comprising:
Receive label-cloud and generate request, wherein carry text set information;
For each label in every section of article in the text set corresponding with described text set information, calculate the weighted value in this label article belonging to it;
Utilize the weighted value of label in each section article corresponding with described text set information, every section of article and described label, generate article-label matrix;
Svd is carried out to described article-label matrix, generates indicative character first matrix of weight of vector in described text set and the second matrix of the weight of indicating label in described proper vector;
The create-rule utilizing described first matrix, the second matrix and pre-set, generating labels cloud.
Preferably, when for each label in every section of article corresponding with described text set information, when calculating the weighted value in this label article belonging to it, following formula is utilized to calculate:
S
(i)(W
k)=[S
source(W
k)-Pos (W
k) * λ (S
source(W
k))] * idf (W
k) * S
attributes(W
k), wherein, described S
(i)(W
k) be a kth label W in i-th section of history article
kthe first weighted value in this history article, described S
source(W
k) be label W
kcarry out source dates, described Pos (W
k) be label W
klocation parameter, described λ (S
source(W
k)) be because of label W
kthe punishment parameter introduced of position, described idf (W
k) be described label W
ksignificance level in internet, described S
attributes(W
k) be described label W
kpart of speech parameter.
Preferably, the label in described utilization each section article corresponding with described text set information, every section of article and the weighted value of described label, generate article-label matrix, comprising:
For every section of article, obtain weighted value in this article and meet the label of the first threshold scope pre-set;
Obtain the union of label described in each;
Each described and concentrated label is utilized to generate article-label matrix, wherein, in described article-label matrix often capable expression one section of article at each described and concentrated label, all articles that a described and concentrated label is corresponding are shown in every list, and the element in this article-label matrix is the weighted value of label.
Preferably, the described create-rule utilizing described first matrix, the second matrix and pre-set, generating labels cloud, comprising:
Obtain each first element meeting in described first matrix and pre-set Second Threshold scope;
For row corresponding with each described first element respectively in described second matrix, obtain in this row and meet label corresponding to each the second element of the 3rd threshold range of pre-setting as the tag element of in label-cloud.
Preferably, also comprise: show the label-cloud be made up of tag element described in each.
Preferably, also comprise: when the quantity of label is greater than preset value in described tag element, delete the part labels in described tag element according to the deletion rule pre-set.
A kind of label-cloud generating apparatus, comprising:
Receiving element, generating request for receiving label-cloud, wherein carrying text set information;
Computing unit, for for each label in every section of article in the text set corresponding with described text set information, calculates the weighted value in this label article belonging to it;
First generation unit, for utilizing the weighted value of label in each section article corresponding with described text set information, every section of article and described label, generates article-label matrix;
Second generation unit, for carrying out svd to described article-label matrix, generates indicative character first matrix of weight of vector in described text set and the second matrix of the weight of indicating label in described proper vector;
3rd generation unit, for the create-rule utilizing described first matrix, the second matrix and pre-set, generating labels cloud.
Preferably, described 3rd generation unit comprises:
Acquiring unit, for obtaining each first element meeting in described first matrix and pre-set Second Threshold scope;
3rd generates subelement, for for row corresponding with each described first element respectively in described second matrix, obtains label corresponding to each the second element of meeting the 3rd threshold range pre-set in this row as the tag element of in label-cloud.
Preferably, also comprise:
Display unit, for showing the label-cloud be made up of tag element described in each.
Preferably, also comprise:
Delete cells, for when the quantity of label is greater than preset value in described tag element, deletes the part labels in described tag element according to the deletion rule pre-set.
The application provides a kind of label-cloud generation method and device, request is generated by receiving the label-cloud carrying text set information, for each label in every section of article in the text set corresponding with text set information, the weighted value calculated in this label article belonging to it generates article-label matrix, and by carrying out svd to article-label matrix, generate indicative character first matrix of weight of vector in text set and the second matrix of the weight of indicating label in proper vector, and then utilize the first matrix, second matrix and the create-rule pre-set realize the generation of label-cloud, the application is by carrying out svd to article-label matrix, and then utilize the matrix generating labels cloud after decomposing, when avoiding the index of label-cloud as article set key content that prior art generates, semantic coverage indicated by each tag element is too wide in range, embody the key content problem not accurately of article set.
Accompanying drawing explanation
In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, be briefly described to the accompanying drawing used required in embodiment or description of the prior art below, apparently, accompanying drawing in the following describes is only embodiments of the invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to the accompanying drawing provided.
A kind of label-cloud that Fig. 1 provides for the embodiment of the present application one generates method flow diagram;
The structural representation of a kind of label-cloud generating apparatus that Fig. 2 provides for the embodiment of the present application two.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, be clearly and completely described the technical scheme in the embodiment of the present invention, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.
Embodiment one:
A kind of label-cloud that Fig. 1 provides for the embodiment of the present application one generates method flow diagram.
As shown in Figure 1, the method comprises:
S101, reception label-cloud generate request, wherein carry text set information.
S102, for each label in every section of article in the text set corresponding with text set information, calculate the weighted value in this label article belonging to it.
In the embodiment of the present application, preferably, text set information correspondence comprises the text set of at least one section of article, and all carry at least one label in every section of article, it is user-defined (as: when preserving certain section of article when the source of this label can be article generation, for it arranges a label for " sweet apple "), also can be through that word segmentation processing obtains (as: by user-defined label " sweet apple " through word segmentation processing, generate two labels " banana " and " apple "/preserve article time, using word more for occurrence number in article as participle label).
Preferably, after receiving the label-cloud carrying text set information, first for each label in every section of article in the text set corresponding with text set information, need calculate the weighted value in this label article belonging to it, wherein, concrete computing formula is as follows:
S
(i)(W
k)=[S
source(W
k)-Pos (W
k) * λ (S
source(W
k))] * idf (W
k) * S
attributes(W
k), wherein, S
(i)(W
k) be a kth label W in i-th section of article
kweighted value in this article, S
source(W
k) be label W
kcarry out source dates, Pos (W
k) be label W
klocation parameter, λ (S
source(W
k)) be because of label W
kthe punishment parameter introduced of position, idf (W
k) be label W
ksignificance level in internet, S
attributes(W
k) be label W
kpart of speech parameter.
In the embodiment of the present application, preferably, S
source(W
k) be label W
kcarry out source dates, wherein, the source of label indicates this label to be customized label/participle label, and preferably, pre-sets the S when label is customized label
source(W
k) when being participle label 8 ~ 20 times of value.
In the embodiment of the present application, preferably, Pos (W
k) be label W
klocation parameter, wherein, the position in each label that the position of label indicates the source in the article of this label belonging to it identical, and preferably, arrange which position in each label that the source in the article of label belonging to it is identical, the Pos (W of this label
k) value is several, as: when carrying 5 labels in certain article, wherein 3 is participle label, and these 3 participle labels are followed successively by " banana ", " apple ", " pear ", so, the Pos (W of label " pear "
k) value is 3.
In the embodiment of the present application, preferably, λ (S
source(W
k)) be because of label W
kthe punishment parameter introduced of position, wherein, punishment parameter is different because the source of label is different, preferably, the λ (S pre-set
source(W
k)) value between 0.08 ~ 0.11, and S
source(W
k)-Pos (W
k) * λ (S
source(W
k)) value be more than or equal to 0.5.
In the embodiment of the present application, preferably, idf (W
k) be label W
ksignificance level in internet, wherein, the process calculating the significance level of certain label is prior art, refers to prior art in detail, is not described in detail at this.
In the embodiment of the present application, preferably, S
attributes(W
k) be label W
kpart of speech parameter, wherein, preferably, the part of speech of label is proper noun, noun, verb, adjective, adverbial word, and when part of speech is proper noun, noun, verb, adjective, adverbial word, is followed successively by S
attributes(W
k) assignment is 10,9,5,4,4.
S103, utilize the weighted value of label in each section article corresponding with text set information, every section of article and label, generation article-label matrix.
In the embodiment of the present application, preferably, when for each label in every section of article in the text set corresponding with text set information, after calculating the weighted value in this label article belonging to it, the weighted value of label in each section article corresponding with text set information, every section of article and label need be utilized, generate article-label matrix, the process of concrete generation article-label matrix is as follows:
1, for every section of article, obtain weighted value in this article and meet the label of the first threshold scope pre-set.
In the embodiment of the present application, preferably, be previously provided with first threshold scope, for every section of article, obtain weighted value in this article and meet the label of the first threshold scope pre-set.
2, the union of each label is obtained.
In the embodiment of the present application, preferably, the label of repetition between article, may be there is, also carry label 1 as article A carries in label 1, article B.
Concrete, when getting after weighted value meets the label of the first threshold scope pre-set for every section of article, for each label in the every section of article got, the union of each label need be obtained.
3, utilize and each label concentrated generate article-label matrix, wherein, in article-label matrix often capable expression one section of article and each label concentrated, every list is shown and all articles corresponding to concentrate label, and the element in this article-label matrix is the weighted value of label.
In the embodiment of the present application, preferably, when after the union getting each label, need be utilized this and each label generation article-label matrix concentrated, wherein, in article-label matrix often capable expression one section of article and each label concentrated, every list is shown and all articles corresponding to concentrate label, and the element in this article-label matrix is the weighted value of label.
Concrete, the article list that the text set corresponding with text set information comprises is { D
1, D
2..., D
n, every section of article D
teach label T
mand this label T
mweighted value W in this section of article
mthe list of labels of composition is { (T
1, w
1) ..., (T
m, w
m), " be more than or equal to θ " by pre-setting first threshold scope, for each label T
mif, its weighted value W
m>=θ, then determine this label T
mfor obtaining result, thus from every section of article D
tin each label T
min filter out that to meet all labels of first threshold scope pre-set be { T
1, w
1) ..., (T
p, w
p), article-label matrix that each tag computation utilizing this to filter out obtains is as follows:
Wherein, n is article quantity, and p is the quantity after union got by each label, and often capable expression one section of article is at each label also concentrated, every list is shown and all articles corresponding to concentrate label, and the element in this article-label matrix is the weighted value of label.
As: when the text set that text set information is corresponding comprises 2 sections of articles, be respectively and carry label A, label B, label C, the article 1 of label D, and carry label A, label B, label C, label D, the article 2 of label E, by for its weighted value in affiliated article of each tag computation, and according to the first threshold scope pre-set, determine that the label meeting this first threshold scope is respectively: the label A (weighted value is 1A) in article 1, label B (weighted value is 1B), label C (weighted value is 1C), and the label A (weighted value is 2A) in article 2, label C (weighted value is 2C), label D (weighted value is 2D), now, article-the label matrix generated is as follows:
Visible, in article-label matrix often capable expression one section of article and concentrate each label, every list is shown and all articles corresponding to concentrate label, and the element in this article-label matrix is the weighted value of label, label B is there is when also concentrating, and article 2 is not when the label also concentrated exists label B, the element of the label B representing article 2 in article-label matrix is arranged to 0.
In the embodiment of the present application, preferably, do not limit the order in the corresponding union label of element in the article-label matrix of this generation in row, inventor can set arbitrarily according to the demand of oneself, and the article-label matrix as generated is:
S104, svd is carried out to article-label matrix, generate first matrix of weight of indicative character vector in text set and the second matrix of the weight of indicating label in proper vector.
In the embodiment of the present application, preferably, carry out svd for the article-label matrix got, generate indicative character first matrix of weight of vector in text set and the second matrix of the weight of indicating label in proper vector.
In the embodiment of the present application, preferably, svd is existing mathematical algorithm, refers to prior art in detail, does not do detailed restriction at this.
S105, the create-rule utilizing the first matrix, the second matrix and pre-set, generating labels cloud.
In the embodiment of the present application, preferably, when carrying out after svd obtains the first matrix and the second matrix to article-label matrix, the create-rule that need utilize the first matrix, the second matrix and pre-set, generating labels cloud, concrete, this process is:
1, each first element meeting in the first matrix and pre-set Second Threshold scope is obtained.
2, for row corresponding with each first element respectively in the second matrix, obtain in this row and meet label corresponding to each the second element of the 3rd threshold range of pre-setting as the tag element of in label-cloud.
In the embodiment of the present application, preferably, the weight of the first matrix indicative character vector in text set, the weight of the second matrix indicating label in proper vector.For certain element in the first matrix, what it indicated is and the weight of this element characteristic of correspondence vector in text set that in the second matrix, indicating label is being row corresponding with this element in the second matrix with the row of the weight in this element characteristic of correspondence vector.
Further, in a kind of label-cloud generation method that the embodiment of the present application provides, also comprise: show the label-cloud be made up of each tag element.
Further, in a kind of label-cloud generation method that the embodiment of the present application provides, also comprise: when the quantity of label is greater than preset value in tag element, delete the part labels in tag element according to the deletion rule pre-set.
In the embodiment of the present application, preferably, when the quantity of the label in certain tag element is greater than preset value, the label in this tag element can be deleted according to the weighted value of each label order from small to large, until the quantity of label meets preset value in this tag element.
The application provides a kind of label-cloud generation method, request is generated by receiving the label-cloud carrying text set information, for each label in every section of article in the text set corresponding with text set information, the weighted value calculated in this label article belonging to it generates article-label matrix, and by carrying out svd to article-label matrix, generate indicative character first matrix of weight of vector in text set and the second matrix of the weight of indicating label in proper vector, and then utilize the first matrix, second matrix and the create-rule pre-set realize the generation of label-cloud, the application is by carrying out svd to article-label matrix, and then utilize the matrix generating labels cloud after decomposing, when avoiding the index of label-cloud as article set key content that prior art generates, semantic coverage indicated by each tag element is too wide in range, embody the key content problem not accurately of article set.
Embodiment two:
The structural representation of a kind of label-cloud generating apparatus that Fig. 2 provides for the embodiment of the present application two.
As shown in Figure 2, this device comprises: the receiving element 1, computing unit 2, first generation unit 3, second generation unit 4 and the 3rd generation unit 5 that are connected successively, wherein:
Receiving element 1, generating request for receiving label-cloud, wherein carrying text set information.
Computing unit 2, for for each label in every section of article in the text set corresponding with text set information, calculates the weighted value in this label article belonging to it.
First generation unit 3, for utilizing the weighted value of label in each section article corresponding with text set information, every section of article and label, generates article-label matrix.
Second generation unit 4, for carrying out svd to article-label matrix, generates indicative character first matrix of weight of vector in text set and the second matrix of the weight of indicating label in proper vector.
3rd generation unit 5, for the create-rule utilizing the first matrix, the second matrix and pre-set, generating labels cloud.
In the embodiment of the present application, preferably, the 3rd generation unit comprises: acquiring unit, for obtaining each first element meeting in the first matrix and pre-set Second Threshold scope; 3rd generates subelement, for for row corresponding with each first element respectively in the second matrix, obtains label corresponding to each the second element of meeting the 3rd threshold range pre-set in this row as the tag element of in label-cloud.
Further, in a kind of label-cloud generating apparatus that the embodiment of the present application provides, also comprise: display unit, for showing the label-cloud be made up of each tag element.
Further, in a kind of label-cloud generating apparatus that the embodiment of the present application provides, also comprise: delete cells, for when the quantity of label is greater than preset value in tag element, delete the part labels in tag element according to the deletion rule pre-set.
The application provides a kind of label-cloud generating apparatus, this device generates request by receiving the label-cloud carrying text set information, for each label in every section of article in the text set corresponding with text set information, the weighted value calculated in this label article belonging to it generates article-label matrix, and by carrying out svd to article-label matrix, generate indicative character first matrix of weight of vector in text set and the second matrix of the weight of indicating label in proper vector, and then utilize the first matrix, second matrix and the create-rule pre-set realize the generation of label-cloud, the application is by carrying out svd to article-label matrix, and then utilize the matrix generating labels cloud after decomposing, when avoiding the index of label-cloud as article set key content that prior art generates, semantic coverage indicated by each tag element is too wide in range, embody the key content problem not accurately of article set.
In this instructions, each embodiment adopts the mode of going forward one by one to describe, and what each embodiment stressed is the difference with other embodiments, between each embodiment identical similar portion mutually see.For device disclosed in embodiment, because it corresponds to the method disclosed in Example, so description is fairly simple, relevant part illustrates see method part.
Below be only the preferred implementation of the application, those skilled in the art understood or realizes the application.To be apparent to one skilled in the art to the multiple amendment of these embodiments, General Principle as defined herein when not departing from the spirit or scope of the application, can realize in other embodiments.Therefore, the application can not be restricted to these embodiments shown in this article, but will meet the widest scope consistent with principle disclosed herein and features of novelty.
Claims (10)
1. a label-cloud generation method, is characterized in that, comprising:
Receive label-cloud and generate request, wherein carry text set information;
For each label in every section of article in the text set corresponding with described text set information, calculate the weighted value in this label article belonging to it;
Utilize the weighted value of label in each section article corresponding with described text set information, every section of article and described label, generate article-label matrix;
Svd is carried out to described article-label matrix, generates indicative character first matrix of weight of vector in described text set and the second matrix of the weight of indicating label in described proper vector;
The create-rule utilizing described first matrix, the second matrix and pre-set, generating labels cloud.
2. method according to claim 1, is characterized in that, when for each label in every section of article corresponding with described text set information, when calculating the weighted value in this label article belonging to it, utilizes following formula to calculate:
S
(i)(W
k)=[S
source(W
k)-Pos (W
k) * λ (S
source(W
k))] * idf (W
k) * S
attributes(W
k), wherein, described S
(i)(W
k) be a kth label W in i-th section of history article
kthe first weighted value in this history article, described S
source(W
k) be label W
kcarry out source dates, described Pos (W
k) be label W
klocation parameter, described λ (S
source(W
k)) be because of label W
kthe punishment parameter introduced of position, described idf (W
k) be described label W
ksignificance level in internet, described S
attributes(W
k) be described label W
kpart of speech parameter.
3. method according to claim 1, is characterized in that, the label in described utilization each section article corresponding with described text set information, every section of article and the weighted value of described label, generate article-label matrix, comprising:
For every section of article, obtain weighted value in this article and meet the label of the first threshold scope pre-set;
Obtain the union of label described in each;
Each described and concentrated label is utilized to generate article-label matrix, wherein, in described article-label matrix often capable expression one section of article at each described and concentrated label, all articles that a described and concentrated label is corresponding are shown in every list, and the element in this article-label matrix is the weighted value of label.
4. method according to claim 1, is characterized in that, the described create-rule utilizing described first matrix, the second matrix and pre-set, and generating labels cloud, comprising:
Obtain each first element meeting in described first matrix and pre-set Second Threshold scope;
For row corresponding with each described first element respectively in described second matrix, obtain in this row and meet label corresponding to each the second element of the 3rd threshold range of pre-setting as the tag element of in label-cloud.
5. method according to claim 4, is characterized in that, also comprises: show the label-cloud be made up of tag element described in each.
6. method according to claim 4, is characterized in that, also comprises: when the quantity of label is greater than preset value in described tag element, delete the part labels in described tag element according to the deletion rule pre-set.
7. a label-cloud generating apparatus, is characterized in that, comprising:
Receiving element, generating request for receiving label-cloud, wherein carrying text set information;
Computing unit, for for each label in every section of article in the text set corresponding with described text set information, calculates the weighted value in this label article belonging to it;
First generation unit, for utilizing the weighted value of label in each section article corresponding with described text set information, every section of article and described label, generates article-label matrix;
Second generation unit, for carrying out svd to described article-label matrix, generates indicative character first matrix of weight of vector in described text set and the second matrix of the weight of indicating label in described proper vector;
3rd generation unit, for the create-rule utilizing described first matrix, the second matrix and pre-set, generating labels cloud.
8. device according to claim 7, is characterized in that, described 3rd generation unit comprises:
Acquiring unit, for obtaining each first element meeting in described first matrix and pre-set Second Threshold scope;
3rd generates subelement, for for row corresponding with each described first element respectively in described second matrix, obtains label corresponding to each the second element of meeting the 3rd threshold range pre-set in this row as the tag element of in label-cloud.
9. device according to claim 8, is characterized in that, also comprises:
Display unit, for showing the label-cloud be made up of tag element described in each.
10. device according to claim 8, is characterized in that, also comprises:
Delete cells, for when the quantity of label is greater than preset value in described tag element, deletes the part labels in described tag element according to the deletion rule pre-set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410534723.XA CN104281690B (en) | 2014-10-11 | 2014-10-11 | A kind of label-cloud generation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410534723.XA CN104281690B (en) | 2014-10-11 | 2014-10-11 | A kind of label-cloud generation method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104281690A true CN104281690A (en) | 2015-01-14 |
| CN104281690B CN104281690B (en) | 2018-01-05 |
Family
ID=52256563
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410534723.XA Active CN104281690B (en) | 2014-10-11 | 2014-10-11 | A kind of label-cloud generation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104281690B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111026832A (en) * | 2019-11-15 | 2020-04-17 | 贝壳技术有限公司 | Method and system for generating articles |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8037066B2 (en) * | 2008-01-16 | 2011-10-11 | International Business Machines Corporation | System and method for generating tag cloud in user collaboration websites |
| US20110296345A1 (en) * | 2010-05-27 | 2011-12-01 | Alcatel-Lucent Usa Inc. | Technique For Determining And Indicating Strength Of An Item In A Weighted List Based On Tagging |
| CN103176961A (en) * | 2013-03-05 | 2013-06-26 | 哈尔滨工程大学 | Transfer learning method based on latent semantic analysis |
| CN103440256A (en) * | 2013-07-26 | 2013-12-11 | 中国科学院深圳先进技术研究院 | Method and device for automatically generating Chinese text label cloud |
-
2014
- 2014-10-11 CN CN201410534723.XA patent/CN104281690B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8037066B2 (en) * | 2008-01-16 | 2011-10-11 | International Business Machines Corporation | System and method for generating tag cloud in user collaboration websites |
| US20110296345A1 (en) * | 2010-05-27 | 2011-12-01 | Alcatel-Lucent Usa Inc. | Technique For Determining And Indicating Strength Of An Item In A Weighted List Based On Tagging |
| CN103176961A (en) * | 2013-03-05 | 2013-06-26 | 哈尔滨工程大学 | Transfer learning method based on latent semantic analysis |
| CN103440256A (en) * | 2013-07-26 | 2013-12-11 | 中国科学院深圳先进技术研究院 | Method and device for automatically generating Chinese text label cloud |
Non-Patent Citations (3)
| Title |
|---|
| SABRI HASSAN.ETC: "SoDA: Dynamic visual analytics of big social data", 《2014 INTERNATIONAL CONFERENCE ON BIG DATA AND SMART COMPUTING 》 * |
| 张林泉等: "基于云计算技术的文本可视化分析", 《成都工业学院学报》 * |
| 黎邦群: "相关关键词与相关图书标签云", 《信息资源建设》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111026832A (en) * | 2019-11-15 | 2020-04-17 | 贝壳技术有限公司 | Method and system for generating articles |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104281690B (en) | 2018-01-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210232760A1 (en) | Optimization techniques for artificial intelligence | |
| EP3485396B1 (en) | Classifying images using machine learning models | |
| CN104615767B (en) | Training method, search processing method and the device of searching order model | |
| CN103207899B (en) | Text recommends method and system | |
| CA3047353A1 (en) | Learning document embeddings with convolutional neural network architectures | |
| CN107256267A (en) | Querying method and device | |
| US20150378986A1 (en) | Context-aware approach to detection of short irrelevant texts | |
| CN104899298A (en) | Microblog sentiment analysis method based on large-scale corpus characteristic learning | |
| CN110162767A (en) | The method and apparatus of text error correction | |
| CN105389307A (en) | Statement intention category identification method and apparatus | |
| CN103810264A (en) | Webpage text classification method based on feature selection | |
| CN102439597A (en) | Parameter deducing method, computing device and system based on potential dirichlet model | |
| CN103678336B (en) | Method and device for identifying entity words | |
| RU2017134978A (en) | INFORMATION SEARCH NAVIGATION METHOD AND DEVICE | |
| CN108959531A (en) | Information search method, device, equipment and storage medium | |
| CN103559313B (en) | Searching method and device | |
| CN104699844B (en) | The method and device of video tab is determined for advertisement | |
| CN104598539A (en) | Internet event hot degree calculation method and terminal | |
| CN104572631A (en) | Training method and system for language model | |
| CN105608075A (en) | Related knowledge point acquisition method and system | |
| CN110110035A (en) | Data processing method and device and computer readable storage medium | |
| CN106649732B (en) | Information pushing method and device | |
| CN110309293A (en) | Text recommended method and device | |
| CN104699819A (en) | Sememe classification method and device | |
| CN104598624A (en) | User class determination method and device for microblog user |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |