CN113688209B - Text semantic network construction method by adjusting keyword dependency relationship - Google Patents
Text semantic network construction method by adjusting keyword dependency relationship Download PDFInfo
- Publication number
- CN113688209B CN113688209B CN202111019581.XA CN202111019581A CN113688209B CN 113688209 B CN113688209 B CN 113688209B CN 202111019581 A CN202111019581 A CN 202111019581A CN 113688209 B CN113688209 B CN 113688209B
- Authority
- CN
- China
- Prior art keywords
- keywords
- frequency
- frequency keywords
- text semantic
- semantic network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/166—Editing, e.g. inserting or deleting
- G06F40/169—Annotation, e.g. comment data or footnotes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention discloses a text semantic network construction method by adjusting the dependency relationship of keywords, which relates to the technical fields of city planning, travel planning, natural language processing and city traffic. Next, a set of high-frequency keywords having a high dependency relationship is extracted from the high-frequency keywords. And traversing comment text data to be analyzed, and establishing connection record relations among the high-frequency keywords in different modes. Finally, constructing different text semantic networks and visualizing the contents of different research topics. According to the invention, important semantic content can be extracted by setting the keyword dependency relationship in the comment file, so that the interference of non-main content on the highlighting of the subject research content is reduced.
Description
Technical Field
The invention relates to the technical fields of city planning, travel planning, urban traffic and complex network modeling, in particular to a text semantic network construction method by adjusting keyword dependency relationship.
Background
The text semantic network construction in natural language has very wide application value in practical application. In the fields of urban planning, travel planning, traffic planning and other industries, the demand for social comment data analysis by using a text semantic network is gradually increasing. The text semantic network mainly aims at establishing a complex network of keyword connection according to the keyword co-occurrence relation in the social comment statement, and further analyzing potential semantic connection characteristics among keywords through various index analysis in the complex network theory. In practical application, the keyword co-occurrence relationship is a technical scheme which is easier to process, and the execution speed of the algorithm is faster and is also easy to be accepted by algorithm users or direct result application users. However, in the use of this technique, some of the existing problems are gradually exposed. Meanwhile, the technology is required to be further improved according to the key requirements in industries such as city planning, travel planning, traffic planning and the like.
The problems of the prior art are: a group of keywords that are not very strong in type or relevance are often considered to have strong connection relationships because they co-occur in comment sentences. For example, in a sentence of social media comments for a high-speed rail station in su state, the keyword "su state station" may be included, but there will often also be a large number of keywords describing the basic features of the high-speed rail station in su state, such as "square", "waiting room", "subway", etc. Likewise, a similar situation may occur for comments from Shanghai stations. When the keywords are too many, if a few words of "Shanghai station" appear in the comments of Suzhou station, the relationship between the two keywords of "Suzhou station" and "Shanghai station" is not significant, because words of "square", "waiting room", "subway" and the like are relatively more frequent in the comments.
This brings about the visual and very important problem that: the node position of the core keywords in the whole semantic network is lower, so that the expression effect of the nodes in the semantic network visualization is affected, and the intention of highlighting a certain theme is not reached.
Disclosure of Invention
The invention aims to solve the technical problem of overcoming the defects of the prior art and providing a text semantic network construction method by adjusting the dependency relationship of keywords. Meanwhile, the scheme of the invention provides a processing technical route of semantic relation among different types of keywords, and achieves the effects of various display and analysis.
The invention adopts the following technical scheme for solving the technical problems:
the text semantic network construction method by adjusting the dependency relationship of the keywords provided by the invention comprises the following steps:
step 1, extracting high-frequency keywords from a group of comment text data to be analyzed;
step 2, extracting a group of high-frequency keywords R with high-degree dependency relationship from the high-frequency keywords, and marking the high-frequency keywords without high-degree dependency relationship as other high-frequency keywords R'; wherein, the high dependency relationship is a preset association relationship;
step 3, traversing comment text data to be analyzed, and establishing a connection record relation R1 between high-frequency keywords;
traversing comment text data to be analyzed, and establishing a new connection record relation R2 according to the R pair of high-frequency keywords;
traversing comment text data to be analyzed, and establishing a new connection record relation R3 between R and R';
wherein, the liquid crystal display device comprises a liquid crystal display device,
traversing comment text data to be analyzed, and establishing a connection record relation R1 between high-frequency keywords; the method comprises the following steps:
step (1), comment text data to be analyzed comprises a plurality of comments, all comments in the comment text data to be analyzed are traversed, one comment is traversed each time, and all high-frequency keywords contained in each comment are identified in the traversing process each time;
step (2), establishing a connection record relation between every two high-frequency keywords aiming at all the high-frequency keywords contained in each comment identified in the step (1);
step (3), if two identified high-frequency keywords appear in the previous traversal process in different comments, the number of connection records between the two high-frequency keywords is increased by 1;
step (4), finally, recording the connection record relation among all the high-frequency keywords as R1;
traversing comment text data to be analyzed, and establishing a new connection record relation R2 according to the R pair of high-frequency keywords; the method comprises the following steps:
step A, in each traversal process: identifying all high-frequency keywords contained in each comment;
step B, aiming at any two different high-frequency keywords identified, if only one high-frequency keyword belongs to R, a connection record relationship is not established between the two high-frequency keywords;
aiming at any two different high-frequency keywords which are identified, if the two high-frequency keywords belong to R or do not belong to R, establishing a connection record relation between every two high-frequency keywords;
step C, if two identified high-frequency keywords appear in previous traversal processing in different comments and meet the requirement of establishing a connection relation in the step B, the number of connection records between the two high-frequency keywords is increased by 1;
step D, finally, recording the connection record relation among all the high-frequency keywords as R2;
traversing comment text data to be analyzed, and establishing a new connection record relation R3 between R and R', wherein the method comprises the following specific steps of:
step a, in each traversal process: identifying all high-frequency keywords contained in each comment;
step b, aiming at any two identified high-frequency keywords, if only one high-frequency keyword belongs to R and the other high-frequency keyword does not belong to R, establishing a connection record relation between the two high-frequency keywords;
for any two identified high-frequency keywords, if both the two high-frequency keywords belong to R or neither the two high-frequency keywords belong to R, a connection record relationship is not established between the two high-frequency keywords;
step c, if two identified high-frequency keywords appear in previous traversal processing in different comments and meet the requirement of establishing a connection relation in the step b, the number of connection records between the two high-frequency keywords is increased by 1;
step d, finally, marking the connection record relation among all the high-frequency keywords as R3;
step 4, constructing three text semantic networks by utilizing a complex network theory;
step 4.1, based on a complex network theory, using high-frequency keywords as nodes of a text semantic network, using R1 as edge connection rules in the text semantic network, constructing the text semantic network, and marking as NET1;
step 4.2, based on a complex network theory, using high-frequency keywords as nodes of a text semantic network, using R2 as edge connection rules in the text semantic network, constructing the text semantic network, and marking as NET2;
and 4.3, constructing a text semantic network based on a complex network theory by taking high-frequency keywords as nodes of the text semantic network and R3 as edge connection rules in the text semantic network, and marking as NET3.
As a further optimization scheme of the text semantic network construction method by adjusting the dependency relationship of the keywords, step 4 is followed by step 5,
step 5, respectively calculating indexes of edges and nodes of the text semantic network by using the three text semantic networks, and visualizing association relations of the text semantic network;
step 5.1, calculating the class index of the network node for the text semantic network NET1;
step 5.2, carrying out community detection division on the network nodes of the text semantic network NET2;
step 5.3, distinguishing the visual display size of the text semantic network node according to the class index calculation result in step 5.1, and carrying out grouping display effect of the text semantic network node according to the community detection division result in step 5.2;
and 5.4, calculating the grade index of the network node for the text semantic network NET3, and visualizing the connection strength relation between the nodes.
As a further optimization scheme of the text semantic network construction method by adjusting the dependency relationship of the keywords, the method 1 specifically comprises the following steps:
step 1.1, extracting keywords from comment text data based on a word segmentation library and a stop word library;
step 1.2, screening a group of high-frequency keywords from the keywords extracted in the step 1.1.
As a further optimization scheme of the text semantic network construction method by adjusting the dependency relationship of the keywords, the sequence of the two high-frequency keywords in the step C in the comment is not used as a judging basis of a new edge connection record.
As a further optimization scheme of the text semantic network construction method by adjusting the dependency relationship of the keywords, the text semantic network constructed in the step 5 is an undirected weighted complex network.
As a further optimization scheme of the text semantic network construction method by adjusting the dependency relationship of the keywords, the class indexes in the step 5.1 and the step 5.4 comprise a degree centrality index and a weight centrality.
Compared with the prior art, the technical scheme provided by the invention has the following technical effects:
(1) The invention provides a text semantic network construction method by adjusting the dependency relationship of keywords, which mainly solves the problem that the expression of core keywords in the text semantic network is not outstanding;
(2) The invention provides a text semantic network construction method by adjusting the dependency relationship of keywords based on a natural language processing technology and a complex network theory, which realizes semantic network expression among various keywords and achieves the purposes of multi-scheme selection and visual display of certain core theme keywords.
Drawings
Fig. 1 is a schematic overall flow diagram of the present invention.
FIG. 2 is a schematic drawing of extracting a set of high frequency keywords with high dependency.
Fig. 3 is a schematic diagram of a connection recording flow between high frequency keywords of R1.
Fig. 4 is a schematic diagram of a connection recording flow between the high frequency keywords of R2.
Fig. 5 is a schematic diagram of a connection recording flow between the high frequency keywords of R3.
FIG. 6 is a schematic diagram of a visualization scheme of a text semantic network; wherein, (a) is a visualization scheme 1, and (b) is a visualization scheme 2.
FIG. 7 is a schematic diagram of a high frequency keyword group visualization effect with high dependency.
FIG. 8 is a schematic diagram of the connection between a high frequency keyword with high dependency and other high frequency keywords.
Detailed Description
The technical scheme of the invention is further described in detail below with reference to the accompanying drawings:
in practical application, a user needs to make a specific relation analysis on a certain group of keywords in the comment keywords, namely, the semantic relation of the group of keywords is considered or hoped to be emphasized through text semantic analysis in advance. Of course, other keywords are not deleted in the analysis, regardless of the other keywords, but are instead combined for observation and analysis. That is, it is necessary to highlight this set of keywords and to retain other keywords. For this purpose, the invention provides the following scheme: the node size of the specific keywords in the semantic network is maintained, and the connection relation between the specific keywords is highlighted. Thus, it is proposed to use 2 text semantic networks to achieve the effect of NET1 and NET2 in the present invention. In order to make up for the lack of expression of the relation between the concerned keywords and other keywords in the first 2 text semantic networks, the invention also proposes to construct a third text semantic network (i.e. NET 3) to solve the problem.
In summary, the scheme of the invention creatively solves an important problem and an urgent need existing in practice through 3 text semantic networks, and provides a better solution for practice and technical research in the fields of urban planning, travel planning, urban traffic and complex network modeling.
In order to solve the problems and needs, the solution of the present invention is: under the requirement of taking the product into consideration to realize automatic calculation, the connection level between the keywords and other keywords is reduced by adjusting the dependency relationship of the keywords, namely, improving the connection level of the important keywords or the keywords to be observed. At the same time, all keywords are to be visually presented together.
The invention provides the following embodiment, namely a text semantic network construction method by adjusting the dependency relationship of keywords, which is mainly characterized by highlighting the purpose of different research topics by adjusting the connection relationship among high-frequency keywords, and specifically comprises the following steps:
step 1) referring to the attached figure 1, extracting high-frequency keywords from a group of comment text data to be analyzed;
step 1.1) extracting keywords from comment text data based on a common word segmentation library and a stop word library;
step 1.2) screening out a group of high-frequency keywords from the keywords extracted in step 1.1.
Step 2) referring to fig. 2, extracting a group of high-frequency keywords with high dependency relationship from the high-frequency keywords, and marking the high-frequency keywords as R; the high-frequency keywords without high dependency relationship are marked as other high-frequency keywords R'; the high dependency relationship is a preset association relationship;
step 3) referring to fig. 3, traversing comment text data to be analyzed, and establishing a connection record relation between high-frequency keywords.
Step 3.1, in each traversal process: identifying all high-frequency keywords contained in each comment;
step 3.2, establishing a connection record relation between every two high-frequency keywords aiming at the identified high-frequency keywords;
step 3.3, if two recognized high-frequency keywords appear in previous traversal processing in different comment sentences, the number of connection records between the two high-frequency keywords is increased by 1;
step 3.4, finally, recording the connection record relation among all the high-frequency keywords as R1;
step 4) referring to fig. 4, traversing comment text data to be analyzed, screening high-frequency keywords R with high dependency relationship, and establishing a new connection record relationship;
step 4.1) during each traversal: identifying all high-frequency keywords contained in each comment;
step 4.2) aiming at any two different high-frequency keywords which are identified, if only one high-frequency keyword belongs to R, a connection record relation is not established between the two high-frequency keywords;
step 4.3) aiming at any two different high-frequency keywords identified, if the two high-frequency keywords belong to R or do not belong to R, establishing a connection record relation between every two high-frequency keywords; the sequence of the two high-frequency keywords in the comment is not used as a judging basis of the new edge connection record;
step 4.4) if two identified high-frequency keywords have appeared in the previous traversal process in different comment sentences and meet the requirements of establishing connection relations in the steps 4.2 and 4.3, the number of connection records between the two high-frequency keywords is increased by 1.
And 4.5) finally, marking the connection record relation among all the high-frequency keywords as R2;
step 5) referring to fig. 5, traversing comment text data to be analyzed, and establishing no connection record for the inside of a high-frequency keyword R with high-degree dependency relationship, and establishing no connection record for the inside of other high-frequency keywords, and establishing a new connection record relationship between the high-frequency keyword R with high-degree dependency relationship and other high-frequency keywords R', wherein the specific steps are as follows:
step 5.1) during each traversal: identifying all high-frequency keywords contained in each comment;
step 5.2) aiming at the identified high-frequency keywords, if only one high-frequency keyword belongs to R and the other high-frequency keyword does not belong to R, establishing a connection record relation between the two high-frequency keywords;
step 5.3) aiming at the identified high-frequency keywords, if both the high-frequency keywords belong to R or neither the high-frequency keywords belong to R, a connection record relationship is not established between the two high-frequency keywords;
step 5.4) if the two identified high-frequency keywords are already present in the previous traversal process in different comment sentences and meet the requirements of establishing the connection relation in the steps 5.2 and 5.3, the number of connection records between the two high-frequency keywords is increased by 1;
and 5.5) finally, marking the connection record relation among all the high-frequency keywords as R3;
the text semantic networks constructed in the step 5 are all undirected weighted complex networks.
Step 6) constructing three text semantic networks by utilizing a complex network theory;
step 6.1) based on a complex network theory, using high-frequency keywords as nodes of a text semantic network, using R1 as an edge connection rule in the text semantic network, constructing the text semantic network, and marking as NET1;
step 6.2) based on the complex network theory, using high-frequency keywords as nodes of a text semantic network, using R2 as edge connection rules in the text semantic network, constructing the text semantic network, and marking as NET2;
step 6.3) based on the complex network theory, using high-frequency keywords as nodes of a text semantic network, using R3 as edge connection rules in the text semantic network, constructing the text semantic network, and marking as NET3;
step 7) referring to fig. 6, using three text semantic networks to respectively calculate indexes of edges and nodes of the text semantic network, and visualizing association relations of the text semantic network;
step 7.1), calculating the class index of the network node for the text semantic network NET1;
step 7.2), carrying out community detection division on network nodes on the text semantic network NET2;
step 7.3) referring to fig. 7, the visual display size of the text semantic network node is distinguished according to the class index calculation result in 6.1, and the grouping display effect of the text semantic network node is carried out according to the community detection division result in 6.2.
Step 7.4) calculating the grade index of the network node for the text semantic network NET3.
The class indexes in step 7.1 and step 7.4 comprise a centrality index and a weighted centrality index.
Referring to fig. 8, the text semantic network NET3 may better reflect the connection relationship between the high frequency keywords having high dependency relationship and other high frequency keywords.
The foregoing is a further detailed description of the invention in connection with the preferred embodiments, and it is not intended that the invention be limited to the specific embodiments described. It will be apparent to those skilled in the art that several simple deductions or substitutions may be made without departing from the spirit of the invention, and these should be considered to be within the scope of the invention.
Claims (6)
1. A text semantic network construction method by adjusting the dependency relationship of keywords is characterized by comprising the following steps:
step 1, extracting high-frequency keywords from a group of comment text data to be analyzed;
step 2, extracting a group of high-frequency keywords R with high-degree dependency relationship from the high-frequency keywords, and marking the high-frequency keywords without high-degree dependency relationship as other high-frequency keywords R'; wherein, the high dependency relationship is a preset association relationship;
step 3, traversing comment text data to be analyzed, and establishing a connection record relation R1 between high-frequency keywords;
traversing comment text data to be analyzed, and establishing a new connection record relation R2 according to the R pair of high-frequency keywords;
traversing comment text data to be analyzed, and establishing a new connection record relation R3 between R and R';
wherein, the liquid crystal display device comprises a liquid crystal display device,
traversing comment text data to be analyzed, and establishing a connection record relation R1 between high-frequency keywords; the method comprises the following steps:
step (1), comment text data to be analyzed comprises a plurality of comments, all comments in the comment text data to be analyzed are traversed, one comment is traversed each time, and all high-frequency keywords contained in each comment are identified in the traversing process each time;
step (2), establishing a connection record relation between every two high-frequency keywords aiming at all the high-frequency keywords contained in each comment identified in the step (1);
step (3), if two identified high-frequency keywords appear in the previous traversal process in different comments, the number of connection records between the two high-frequency keywords is increased by 1;
step (4), finally, recording the connection record relation among all the high-frequency keywords as R1;
traversing comment text data to be analyzed, and establishing a new connection record relation R2 according to the R pair of high-frequency keywords; the method comprises the following steps:
step A, in each traversal process: identifying all high-frequency keywords contained in each comment;
step B, aiming at any two different high-frequency keywords identified, if only one high-frequency keyword belongs to R, a connection record relationship is not established between the two high-frequency keywords;
aiming at any two different high-frequency keywords which are identified, if the two high-frequency keywords belong to R or do not belong to R, establishing a connection record relation between every two high-frequency keywords;
step C, if two identified high-frequency keywords appear in previous traversal processing in different comments and meet the requirement of establishing a connection relation in the step B, the number of connection records between the two high-frequency keywords is increased by 1;
step D, finally, recording the connection record relation among all the high-frequency keywords as R2;
traversing comment text data to be analyzed, and establishing a new connection record relation R3 between R and R', wherein the method comprises the following specific steps of:
step a, in each traversal process: identifying all high-frequency keywords contained in each comment;
step b, aiming at any two identified high-frequency keywords, if only one high-frequency keyword belongs to R and the other high-frequency keyword does not belong to R, establishing a connection record relation between the two high-frequency keywords;
for any two identified high-frequency keywords, if both the two high-frequency keywords belong to R or neither the two high-frequency keywords belong to R, a connection record relationship is not established between the two high-frequency keywords;
step c, if two identified high-frequency keywords appear in previous traversal processing in different comments and meet the requirement of establishing a connection relation in the step b, the number of connection records between the two high-frequency keywords is increased by 1;
step d, finally, marking the connection record relation among all the high-frequency keywords as R3;
step 4, constructing three text semantic networks by utilizing a complex network theory;
step 4.1, based on a complex network theory, using high-frequency keywords as nodes of a text semantic network, using R1 as edge connection rules in the text semantic network, constructing the text semantic network, and marking as NET1;
step 4.2, based on a complex network theory, using high-frequency keywords as nodes of a text semantic network, using R2 as edge connection rules in the text semantic network, constructing the text semantic network, and marking as NET2;
and 4.3, constructing a text semantic network based on a complex network theory by taking high-frequency keywords as nodes of the text semantic network and R3 as edge connection rules in the text semantic network, and marking as NET3.
2. The text semantic network construction method by adjusting keyword dependency according to claim 1, wherein step 4 is followed by step 5,
step 5, respectively calculating indexes of edges and nodes of the text semantic network by using the three text semantic networks, and visualizing association relations of the text semantic network;
step 5.1, calculating the class index of the network node for the text semantic network NET1;
step 5.2, carrying out community detection division on the network nodes of the text semantic network NET2;
step 5.3, distinguishing the visual display size of the text semantic network node according to the class index calculation result in step 5.1, and carrying out grouping display effect of the text semantic network node according to the community detection division result in step 5.2;
and 5.4, calculating the grade index of the network node for the text semantic network NET3, and visualizing the connection strength relation between the nodes.
3. The text semantic network construction method by adjusting the dependency relationship of keywords according to claim 1, wherein in the step 1, the following is specific:
step 1.1, extracting keywords from comment text data based on a word segmentation library and a stop word library;
step 1.2, screening a group of high-frequency keywords from the keywords extracted in the step 1.1.
4. The text semantic network construction method by adjusting the dependency relationship of keywords according to claim 1, wherein the sequence of the two high-frequency keywords in the comment in the step C is not used as a basis for judging a new edge connection record.
5. The text semantic network construction method by adjusting keyword dependency relationship according to claim 1, wherein the text semantic networks constructed in step 5 are all undirected weighted complex networks.
6. The text semantic network construction method by adjusting keyword dependency according to claim 2, wherein the class level indicators in step 5.1 and step 5.4 include a degree-centering indicator and a weight-centering indicator.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111019581.XA CN113688209B (en) | 2021-09-01 | 2021-09-01 | Text semantic network construction method by adjusting keyword dependency relationship |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111019581.XA CN113688209B (en) | 2021-09-01 | 2021-09-01 | Text semantic network construction method by adjusting keyword dependency relationship |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113688209A CN113688209A (en) | 2021-11-23 |
CN113688209B true CN113688209B (en) | 2023-08-25 |
Family
ID=78584639
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111019581.XA Active CN113688209B (en) | 2021-09-01 | 2021-09-01 | Text semantic network construction method by adjusting keyword dependency relationship |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113688209B (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20100060734A (en) * | 2008-11-28 | 2010-06-07 | 한국과학기술원 | System for visualization of patent information by forming the keyword based semantic network and method therefor |
CN105893611A (en) * | 2016-04-27 | 2016-08-24 | 南京邮电大学 | Method for establishing interest theme semantic network facing to social networking services |
CN106776678A (en) * | 2016-11-02 | 2017-05-31 | 四川用联信息技术有限公司 | Search engine optimization technology is realized in new keyword optimization |
CN107679084A (en) * | 2017-08-31 | 2018-02-09 | 平安科技(深圳)有限公司 | Cluster labels generation method, electronic equipment and computer-readable recording medium |
CN111625622A (en) * | 2020-04-28 | 2020-09-04 | 北京中科智加科技有限公司 | Domain ontology construction method and device, electronic equipment and storage medium |
-
2021
- 2021-09-01 CN CN202111019581.XA patent/CN113688209B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20100060734A (en) * | 2008-11-28 | 2010-06-07 | 한국과학기술원 | System for visualization of patent information by forming the keyword based semantic network and method therefor |
CN105893611A (en) * | 2016-04-27 | 2016-08-24 | 南京邮电大学 | Method for establishing interest theme semantic network facing to social networking services |
CN106776678A (en) * | 2016-11-02 | 2017-05-31 | 四川用联信息技术有限公司 | Search engine optimization technology is realized in new keyword optimization |
CN107679084A (en) * | 2017-08-31 | 2018-02-09 | 平安科技(深圳)有限公司 | Cluster labels generation method, electronic equipment and computer-readable recording medium |
CN111625622A (en) * | 2020-04-28 | 2020-09-04 | 北京中科智加科技有限公司 | Domain ontology construction method and device, electronic equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN113688209A (en) | 2021-11-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Olmeda-Gómez et al. | Co-word analysis and thematic landscapes in Spanish information science literature, 1985–2014 | |
CN103544255B (en) | Text semantic relativity based network public opinion information analysis method | |
CN106339450B (en) | Index method of tree-shaped data | |
Peris et al. | The evolution of the systems of cities literature since 1995: Schools of thought and their interaction | |
CN110674311A (en) | Knowledge graph-based power asset heterogeneous data fusion method | |
CN113065003B (en) | Knowledge graph generation method based on multiple indexes | |
CN110457404A (en) | Social media account-classification method based on complex heterogeneous network | |
Alomari et al. | Sentiment analysis of Arabic tweets for road traffic congestion and event detection | |
CN105068991A (en) | Big data based public sentiment discovery method | |
JP5907393B2 (en) | Complex predicate template collection device and computer program therefor | |
CN105893611A (en) | Method for establishing interest theme semantic network facing to social networking services | |
CN105630884A (en) | Geographic position discovery method for microblog hot event | |
CN111897914A (en) | Entity information extraction and knowledge graph construction method for field of comprehensive pipe gallery | |
CN109582761A (en) | A kind of Chinese intelligent Answer System method of the Words similarity based on the network platform | |
CN115545468A (en) | Audit risk measurement method based on knowledge graph | |
Nedeljković et al. | Building a construction project key-phrase network from unstructured text documents | |
Ferret | Typing relations in distributional thesauri | |
CN113688209B (en) | Text semantic network construction method by adjusting keyword dependency relationship | |
Chen et al. | Novelty paper recommendation using citation authority diffusion | |
CN109033370A (en) | A kind of method and device that searching similar shop, the method and device of shop access | |
CN103488743B (en) | Page element extraction method and page element extraction system | |
CN116467291A (en) | Knowledge graph storage and search method and system | |
Bernabé-Moreno et al. | Leveraging localized social media insights for industry early warning systems | |
Zheng et al. | Architecture Descriptions Analysis Based on Text Mining and Crawling Technology | |
CN114638225A (en) | Automatic keyword extraction method based on scientific and technological literature graph network |
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 |