CN113505233B - Extraction method of ecological civilized geographic knowledge based on open domain - Google Patents
Extraction method of ecological civilized geographic knowledge based on open domain Download PDFInfo
- Publication number
- CN113505233B CN113505233B CN202110632026.8A CN202110632026A CN113505233B CN 113505233 B CN113505233 B CN 113505233B CN 202110632026 A CN202110632026 A CN 202110632026A CN 113505233 B CN113505233 B CN 113505233B
- Authority
- CN
- China
- Prior art keywords
- geographic
- time
- ecological
- space
- civilized
- 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
- 238000000605 extraction Methods 0.000 title claims abstract description 51
- 238000000034 method Methods 0.000 claims abstract description 25
- 238000004458 analytical method Methods 0.000 claims abstract description 19
- 238000002372 labelling Methods 0.000 claims abstract description 17
- 239000002245 particle Substances 0.000 claims abstract description 14
- 238000010276 construction Methods 0.000 claims description 6
- 238000001514 detection method Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 239000002689 soil Substances 0.000 claims description 4
- 238000012876 topography Methods 0.000 claims description 4
- 238000009827 uniform distribution Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 11
- 230000009286 beneficial effect Effects 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 4
- 239000011435 rock Substances 0.000 description 4
- 230000004927 fusion Effects 0.000 description 3
- 230000014509 gene expression Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000001556 precipitation Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 230000002123 temporal effect Effects 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
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/27—Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Abstract
The invention relates to the field of ecological civilization, and provides an extraction method of ecological civilization geographic knowledge based on an open domain, wherein the method acquires multi-source heterogeneous network data by adopting a distributed network data acquisition technology; text space-time analysis is carried out on the multi-source heterogeneous network data based on space-time scene event particles, and ecological civilization geographic space-time scene events, corresponding time and position attributes are extracted; detecting an ecological civilized geographic topic based on latent semantic analysis; constructing a labeling corpus based on a crowdsourcing mode; and constructing an ecological civilized geographic space-time event database according to the ecological civilized geographic space-time scene event, the corresponding time and position attribute, the theme and the corpus. The method mainly solves the problem of low extraction performance of the implicit relation extracted by the knowledge entity in the ecological civilized geographic field, and is beneficial to improving the extraction performance of the implicit relation and the accuracy of the extraction result.
Description
Technical Field
The invention relates to the technical field of ecological civilization construction, in particular to an extraction method of ecological civilization geographic knowledge based on an open domain.
Background
Knowledge graph (knowledgegraph) is a branch of the artificial intelligence field, and is a mode of most important knowledge representation in the big data era. At present, the knowledge graph has been applied in intelligent searching, deep question and answer, social networks and some vertical industries, and becomes a power source for supporting the development of the application. For example, the knowledge graph of google sequentially integrates public resources such as Wikipedia, CIA world overview and a large amount of semantic data collected and arranged from other websites, and big social service sites such as Bing Search and Facebook, twitter of Microsoft achieve a cooperation agreement, so that the method has remarkable advantages in the aspects of collecting and customizing personalized contents of users. Social networking sites Facebook introduced Graph Search products in 2013, and the core technology is to link people, places, things and the like together through a knowledge Graph, and support accurate natural language query in an intuitive way, such as input query formula: "restaurant I friends like" live in New York and like basketball and Chinese movies "etc., the knowledge graph will help users find people, photos, places and interests, etc. that are most relevant to themselves in a huge social network.
The knowledge graph core technology comprises knowledge extraction, knowledge representation, knowledge fusion and knowledge reasoning technology. The knowledge extraction is mainly aimed at open link data, available knowledge units are extracted through an automatic or semi-automatic technology, and a series of high-quality fact expressions are formed and stored in a knowledge graph based on the available knowledge units. The knowledge unit mainly comprises 3 knowledge elements of entity, relationship and attribute, so the knowledge extraction mainly comprises 3 processes of entity extraction, relationship extraction and attribute extraction. Entity extraction refers to automatically identifying named entities from raw data corpus, and is one step of most basic and key in knowledge extraction. The entity extraction method is divided into 3 types, including a rule and dictionary based method, a statistical machine learning based method and an open domain oriented extraction method. The relation extraction goal is to solve the problem of semantic links between entities. An open domain oriented information extraction framework (open information extraction, OIE) is pioneering, but the implicit relationship extraction performance of the OIE method entity is low. Yang Bo et al (2014) propose a deep implicit relation extraction method based on markov logic network and ontology reasoning, but the accuracy of the extraction result is lower. Attribute extraction is primarily directed to entities, from which a complete sketch of an entity can be formed. The attribute extraction method based on the rule and the heuristic algorithm can automatically extract corresponding attribute names and attribute values from the semi-structured web pages of Wikipedia and WordNet, and can be further expanded into a set of ontology knowledge base.
Disclosure of Invention
Therefore, in order to improve the extraction performance of the implicit relation and the accuracy of the extraction result, the invention provides an extraction method of ecological civilized geographic knowledge based on an open domain.
Specifically, the method is realized mainly by the following technical scheme:
an extraction method of ecological civilized geographic knowledge based on an open domain comprises the following steps:
acquiring multi-source heterogeneous network data by adopting a distributed network data acquisition technology;
text space-time analysis is carried out on the multi-source heterogeneous network data based on space-time scene event particles, and ecological civilization geographic space-time scene events, corresponding time and position attributes are extracted;
detecting an ecological civilized geographic topic based on latent semantic analysis;
constructing a labeling corpus based on a crowdsourcing mode;
and constructing an ecological civilized geographic space-time event database according to the ecological civilized geographic space-time scene event, the corresponding time and position attribute, the theme and the corpus.
Preferably, the ecological civilized geographic theme comprises: topography, climate, hydrology, soil, ecological environment, natural disasters, socioeconomic, population and folk culture, major engineering construction and ecological civilization treatment.
Preferably, the text spatiotemporal analysis is performed on the multi-source heterogeneous network data based on spatiotemporal scene event particles, and specifically comprises the following steps: and based on the time and position attribute extraction and the space-time scene cutting of the context, cutting the multi-source heterogeneous network data into a space-time event set which is formed by taking scene particle events as units.
Preferably, the ecological civilization geographic theme and the theme type are detected by using an LDA and LabeledLDA theme-based model.
Preferably, an ecological civilized geographic space-time event automatic detection and labeling online system is utilized to establish a labeling corpus and an ecological civilized geographic subject training model in a crowdsourcing mode.
Preferably, extracting the ecological civilized geographic spatiotemporal scene event includes extracting an entity of the ecological civilized geographic spatiotemporal scene event, specifically including: extracting an encyclopedia knowledge map of a professional academic tool book; or identifying the document abstract based on the BiLSTM deep neural network and the conditional random field; or, the topic-based classification feature cluster recognition is oriented to an open domain.
Preferably, the entity extracted is fused into a geographic phenomenon or process.
Preferably, a uniform distribution model is constructed for space-time alignment, and the corresponding time and position attribute description granularity coarser ecological civilization geographic space-time scene events are projected to a space-time cube.
The invention acquires multi-source heterogeneous network data by adopting a distributed network data acquisition technology; text space-time analysis is carried out on the multi-source heterogeneous network data based on space-time scene event particles, and ecological civilization geographic space-time scene events, corresponding time and position attributes are extracted; detecting an ecological civilized geographic topic based on latent semantic analysis; constructing a labeling corpus based on a crowdsourcing mode; and constructing an ecological civilized geographic space-time event database according to the ecological civilized geographic space-time scene event, the corresponding time and position attribute, the theme and the corpus. The method mainly solves the problem that the extraction performance of the implicit relation extracted by the knowledge entity in the ecological civilized geographic field is low, and is beneficial to improving the extraction performance of the implicit relation and the accuracy of an extraction result.
Therefore, in order to improve the extraction performance of the implicit relation and the accuracy of the extraction result, the invention provides an extraction method of ecological civilized geographic knowledge based on an open domain.
Specifically, the method is realized mainly by the following technical scheme:
an extraction method of ecological civilized geographic knowledge based on an open domain comprises the following steps:
acquiring multi-source heterogeneous network data by adopting a distributed network data acquisition technology;
text space-time analysis is carried out on the multi-source heterogeneous network data based on space-time scene event particles, and ecological civilization geographic space-time scene events, corresponding time and position attributes are extracted;
detecting an ecological civilized geographic topic based on latent semantic analysis;
constructing a labeling corpus based on a crowdsourcing mode;
and constructing an ecological civilized geographic space-time event database according to the ecological civilized geographic space-time scene event, the corresponding time and position attribute, the theme and the corpus.
Preferably, the ecological civilized geographic theme comprises: topography, climate, hydrology, soil, ecological environment, natural disasters, socioeconomic, population and folk culture, major engineering construction and ecological civilization treatment.
Preferably, the text spatiotemporal analysis is performed on the multi-source heterogeneous network data based on spatiotemporal scene event particles, and specifically comprises the following steps: and based on the time and position attribute extraction and the space-time scene cutting of the context, cutting the multi-source heterogeneous network data into a space-time event set which is formed by taking scene particle events as units.
Preferably, the ecological civilization geographic theme and the theme type are detected by using an LDA and LabeledLDA theme-based model.
Preferably, an ecological civilized geographic space-time event automatic detection and labeling online system is utilized to establish a labeling corpus and an ecological civilized geographic subject training model in a crowdsourcing mode.
Preferably, extracting the ecological civilized geographic spatiotemporal scene event includes extracting an entity of the ecological civilized geographic spatiotemporal scene event, specifically including: extracting an encyclopedia knowledge map of a professional academic tool book; or identifying the document abstract based on the BiLSTM deep neural network and the conditional random field; or, the topic-based classification feature cluster recognition is oriented to an open domain.
Preferably, the entity extracted is fused into a geographic phenomenon or process.
Preferably, a uniform distribution model is constructed for space-time alignment, and the corresponding time and position attribute description granularity coarser ecological civilization geographic space-time scene events are projected to a space-time cube.
The invention acquires multi-source heterogeneous network data by adopting a distributed network data acquisition technology; text space-time analysis is carried out on the multi-source heterogeneous network data based on space-time scene event particles, and ecological civilization geographic space-time scene events, corresponding time and position attributes are extracted; detecting an ecological civilized geographic topic based on latent semantic analysis; constructing a labeling corpus based on a crowdsourcing mode; and constructing an ecological civilized geographic space-time event database according to the ecological civilized geographic space-time scene event, the corresponding time and position attribute, the theme and the corpus. The method mainly solves the problem that the extraction performance of the implicit relation extracted by the knowledge entity in the ecological civilized geographic field is low, and is beneficial to improving the extraction performance of the implicit relation and the accuracy of an extraction result.
Drawings
FIG. 1 is a logic flow diagram of an extraction method of ecological civilized geographic knowledge based on an open domain provided in an embodiment of the present invention;
FIG. 2 is a schematic diagram of analysis results of spatiotemporal scene event particles provided in an embodiment of the present invention;
FIG. 3 is a diagram of an IncadentNet network provided in an embodiment of the present invention;
fig. 4 is a schematic view of a spatiotemporal cube provided in an embodiment of the invention.
Description of the embodiments
In order to more clearly understand the core idea of the present invention, a detailed description will be given below with reference to the accompanying drawings.
The ubiquitous network contains massive data reflecting the recent modern history and the current ecological civilization geographic evolution process of China. However, because the ubiquitous network data lacks uniform semi-structural characteristics similar to the encyclopedic knowledge web pages, the acquisition of the ubiquitous network data based on knowledge can only adopt an open domain extraction mode. The ecological civilization geographic features are often embodied as geographic phenomena or processes, are space-time procedural results of the combined action of various elements, so that the ecological civilization geographic network information is often unidentified, the text expression form is wide, the relevance is implicit and fuzzy, the extraction difficulty of a knowledge entity is high, the results are greatly influenced by factors such as scale, angle and subject classification, and therefore, the results are difficult to directly process by using a traditional method. Therefore, the invention adopts a distributed network data acquisition technology to form the acquisition capability of multi-source heterogeneous network data business, realizes automatic detection and time and position attribute extraction of events based on text space-time analysis of event particles of space-time scene, detects ecological civilized geographic subjects (Topic) based on potential semantic analysis and builds a corpus marked based on a crowdsourcing mode, and realizes event attribute extraction based on part-of-speech analysis, thereby forming a space-time event 3W (time when, position where and event content what) element extraction scheme so as to conveniently build a massive ecological civilized geographic space-time event database.
An extraction method of ecological civilized geographic knowledge based on an open domain, as shown in fig. 1, specifically comprises the following steps:
s1, acquiring multi-source heterogeneous network data by adopting a distributed network data acquisition technology.
It should be noted that, the data may be acquired through a multi-channel-based network data acquisition platform, for example, a news portal (including, for example, newfashioned, fox searching, tencer, people net and Xinhua net), a government/industry agency website (including, for example, central, provincial, regional and county level four government websites and country, environment, planning, agriculture, forestry, animal husbandry, industry, population, disaster emergency and other industry management agency websites), a microblog social website, a community forum website and the like.
And S2, carrying out text space-time analysis on the multi-source heterogeneous network data based on space-time scene event particles, and extracting ecological civilization geographic space-time scene events, corresponding time and position attributes.
The method for extracting time and position attributes and cutting space-time scenes based on the context cuts the acquired multi-source heterogeneous network data file into a space-time event (spatio-temporal) set consisting of scene particle events, wherein the space-time event consists of three elements including time (where, including standard time and original time description), position (where, including longitude and latitude coordinates, place name address information and space granularity) and content (content), namely, when and where things related to geographic phenomena, geographic processes and ecological civilization construction occur.
As shown in fig. 2, for example: mud-rock flow occurs in the area A and other places in 7 and 29 1995, and 32 households suffer from disasters, rushing down between houses 160, and 36 ditches of 484 mu of cultivated land and earth-rock dam. The 7 months and 5 days after the year, mud-rock flow occurs in the region B, 1 person dies, and the house 4 and part of cultivated land and trees are destroyed. Two spatiotemporal events can be formed with relative independence.
And S3, detecting the ecological civilized geographic theme based on latent semantic analysis.
In a preferred embodiment, the topic and topic type of the ecological civilized geographic spatiotemporal event content are detected based on LDA and labelelda topic models so as to achieve the purpose of extraction of "what (what)" by combining automation and semi-automation.
In a preferred embodiment, the ecological civilized geographic theme comprises: topography, climate, hydrology, soil, ecological environment, natural disasters, socioeconomic, population and folk culture, major engineering construction and ecological civilization treatment.
Examples are, for example: "region C,3 families, 21 families, except that the workers leave home and survive, the rest 19 people all suffer, wherein one 10 families all die. "; "casualties"; "region D is flushed 9 people by debris flow, wherein one 8 people flush 7 people. "; "mud-rock flow &personnelcasualties".
S4, constructing a labeling corpus based on a crowdsourcing mode.
In a preferred embodiment, a labeling corpus and an ecological civilized geographic subject training model are established by an ecological civilized geographic space-time event automatic detection and labeling online system in a crowdsourcing mode.
The ecological civilization geographic space-time event automatic detection and labeling online system establishes a labeling corpus and a theme training model by utilizing a crowdsourcing mode. The basic idea of the two mode fusion is as follows: firstly, using a hierarchical knowledge classification system as a starting training corpus, and detecting the space-time event content by using LabeledLDA; the data which fail to be detected are further iterated by using LDA; after eliminating the data of the clear theme attribute, manually marking in a crowdsourcing mode by utilizing an online marking system; and finally obtaining the theme attribute of the event content and the newly added theme type.
S5, constructing an ecological civilized geographic space-time event database according to the ecological civilized geographic space-time scene event, the corresponding time and position attribute, the theme and the corpus.
The multi-source heterogeneous network data document is divided into a space-time event set according to a space-time scene. The spatio-temporal events include 3W elements, so that links between events can be established based on the similarity of the 3W elements, a link relationship between events is formed, and further, a link relationship between events and documents, and a link relationship between documents are formed. Three links can be used to construct a network diagram, known as an encidentnet, as shown in fig. 3.
The event similarity calculation adopts cosine distance, and utilizes multiple dimensions such as time, position, theme and keyword set to construct the link relation between events. Meanwhile, by utilizing an ecological civilized geographic knowledge system, according to the classification of the space-time event subject, the ecological civilized geographic knowledge fusion approach is formed by automatically merging and fusing, for example, according to the body model of flood disasters, elements such as precipitation, casualties, economic loss degree, disaster relief measures and the like which are distributed in different documents and related to the space-time event are integrated into a whole.
In a preferred embodiment, extracting the ecological civilized geographic spatiotemporal scene event includes extracting an entity of the ecological civilized geographic spatiotemporal scene event, specifically including: extracting an encyclopedia knowledge map of a professional academic tool book; or identifying the document abstract based on the BiLSTM deep neural network and the conditional random field; or, the topic-based classification feature cluster recognition is oriented to an open domain.
In a preferred embodiment, the entities extracted are fused into a geographic phenomenon or process.
In a preferred embodiment, a uniformly distributed model is constructed for space-time alignment, and corresponding time and position attribute description coarse-granularity ecological civilization geographic space-time scene events are projected to a space-time cube.
In view of the problem that the space-time description of the event is inconsistent, space-time semantic disambiguation processing is needed, and space-time quantization granularity is enhanced. For example, "great earthquake in the winter in 2008", "great earthquake in the Wen in 2008", "great earthquake in the Sichuan in the 5 th month and 12 th year", etc., have the same spatiotemporal semantics under certain context constraints, namely, "earthquake occurring in the Wen in the Sichuan in the 5 th month and 12 th year in 2008". Thus, a spatio-temporal alignment method is constructed based on a uniformly distributed model, projecting events with coarser temporal, positional and topic granularity onto a relatively precise spatio-temporal cube, as shown in FIG. 4. The method is beneficial to eliminating the problems of ambiguity and unknown pointing of time, position, theme classification and other dimensions.
The foregoing has outlined rather broadly the more detailed description of embodiments of the invention, wherein the principles and embodiments of the invention are explained in detail using specific examples, the description of the embodiments being merely intended to facilitate an understanding of the general principles of the invention; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in accordance with the ideas of the present invention, the present description should not be construed as limiting the present invention in view of the above.
Claims (3)
1. The extraction method of the ecological civilized geographic knowledge based on the open domain is characterized by comprising the following steps:
acquiring multi-source heterogeneous network data by adopting a multi-channel network data acquisition platform;
based on the time and position attribute extraction and space-time scene cutting of the context, cutting the multi-source heterogeneous network data into a space-time event set formed by taking scene particle events as units, and extracting ecological civilization geographic space-time scene events and corresponding time and position attributes;
detecting an ecological civilized geographic topic based on latent semantic analysis;
constructing a labeling corpus based on a crowdsourcing mode;
constructing an ecological civilized geographic space-time event database according to the ecological civilized geographic space-time scene event, the corresponding time and position attribute, the theme and the corpus;
constructing a uniform distribution model for space-time alignment, and projecting corresponding ecological civilized geographic space-time scene events with thicker time and position attribute description granularity to a space-time cube;
wherein, ecological civilized geographic theme includes: topography, climate, hydrology, soil, ecological environment, natural disasters, socioeconomic, population, folk culture, major engineering construction and ecological civilization management;
extracting the ecological civilized geographic space-time scene event comprises extracting an entity of the ecological civilized geographic space-time scene event, and specifically comprises the following steps: extracting an encyclopedia knowledge map of a professional academic tool book; or identifying the document abstract based on the BiLSTM deep neural network and the conditional random field; or, the entity is fused into geographic phenomenon or process based on topic classification feature cluster recognition facing the open domain.
2. The method for extracting the ecological civilized geographic knowledge based on the open domain according to claim 1, wherein the ecological civilized geographic topics and topic types are detected based on the LDA and labelelda topic models.
3. The method for extracting the open domain-based ecological civilized geographic knowledge as in claim 1, further comprising: and establishing a labeling corpus and an ecological civilized geographic subject training model by using a crowdsourcing mode through an automatic detection and labeling online system of the ecological civilized geographic space-time events.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110632026.8A CN113505233B (en) | 2021-06-07 | 2021-06-07 | Extraction method of ecological civilized geographic knowledge based on open domain |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110632026.8A CN113505233B (en) | 2021-06-07 | 2021-06-07 | Extraction method of ecological civilized geographic knowledge based on open domain |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113505233A CN113505233A (en) | 2021-10-15 |
CN113505233B true CN113505233B (en) | 2023-11-21 |
Family
ID=78009056
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110632026.8A Active CN113505233B (en) | 2021-06-07 | 2021-06-07 | Extraction method of ecological civilized geographic knowledge based on open domain |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113505233B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114742055A (en) * | 2022-03-29 | 2022-07-12 | 北京感易智能科技有限公司 | Data processing method, data processing apparatus, electronic device, medium, and program product |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109918478A (en) * | 2019-02-26 | 2019-06-21 | 北京悦图遥感科技发展有限公司 | The method and apparatus of knowledge based map acquisition geographic products data |
CN110472066A (en) * | 2019-08-07 | 2019-11-19 | 北京大学 | A kind of construction method of urban geography semantic knowledge map |
CN112069246A (en) * | 2020-09-08 | 2020-12-11 | 天津大学 | Analysis method for event evolution process integration in physical world and network world |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11222052B2 (en) * | 2011-02-22 | 2022-01-11 | Refinitiv Us Organization Llc | Machine learning-based relationship association and related discovery and |
US10303999B2 (en) * | 2011-02-22 | 2019-05-28 | Refinitiv Us Organization Llc | Machine learning-based relationship association and related discovery and search engines |
-
2021
- 2021-06-07 CN CN202110632026.8A patent/CN113505233B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109918478A (en) * | 2019-02-26 | 2019-06-21 | 北京悦图遥感科技发展有限公司 | The method and apparatus of knowledge based map acquisition geographic products data |
CN110472066A (en) * | 2019-08-07 | 2019-11-19 | 北京大学 | A kind of construction method of urban geography semantic knowledge map |
CN112069246A (en) * | 2020-09-08 | 2020-12-11 | 天津大学 | Analysis method for event evolution process integration in physical world and network world |
Non-Patent Citations (5)
Title |
---|
Linking OpenStreetMap with knowledge graphs — Link discovery for schema-agnostic volunteered geographic information;Nicolas Tempelmeier 等;《Future Generation Computer Systems》;349-364 * |
一种准确而高效的领域知识图谱构建方法;杨玉基 等;《软件学报》;2931-2947 * |
中文文本蕴含气象灾害事件信息多模型融合抽取方法;胡段牧 等;《地球信息科学学报》;2342-2355 * |
地质知识图谱标准化模型研究;袁满 等;《吉林大学学报(信息科学版) 》;215-222 * |
面向化工领域的实体关系抽取技术研究;杜坤钰;《中国优秀硕士学位论文全文数据库工程科技Ⅰ辑》;B016-122,正文第3-6章 * |
Also Published As
Publication number | Publication date |
---|---|
CN113505233A (en) | 2021-10-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Resch et al. | Combining machine-learning topic models and spatiotemporal analysis of social media data for disaster footprint and damage assessment | |
CN110472066B (en) | Construction method of urban geographic semantic knowledge map | |
CN107205016B (en) | Retrieval method of Internet of things equipment | |
Ghahremanlou et al. | Geotagging twitter messages in crisis management | |
CN109033358A (en) | News Aggreagation and the associated method of intelligent entity | |
CN113505234B (en) | Construction method of ecological civilized geographic knowledge graph | |
CN103514234A (en) | Method and device for extracting page information | |
CN111899089A (en) | Enterprise risk early warning method and system based on knowledge graph | |
CN111831794A (en) | Knowledge map-based construction method for knowledge question-answering system in comprehensive pipe gallery industry | |
CN109783484A (en) | The construction method and system of the data service platform of knowledge based map | |
Jindal et al. | Construction of domain ontology utilizing formal concept analysis and social media analytics | |
CN112288247A (en) | Soil heavy metal risk identification method based on space interaction relation | |
Shen et al. | Information retrieval of a disaster event from cross-platform social media | |
CN112434168A (en) | Knowledge graph construction method and fragmentized knowledge generation method based on library | |
CN111966787A (en) | Intelligent fishery question-answering robot construction method based on knowledge graph | |
Bahrehdar et al. | Description and characterization of place properties using topic modeling on georeferenced tags | |
CN113505233B (en) | Extraction method of ecological civilized geographic knowledge based on open domain | |
Ireson | Local community situational awareness during an emergency | |
Sukel et al. | Multimodal classification of urban micro-events | |
CN109087223A (en) | A kind of educational resource model building method based on ontology | |
Jing et al. | Integration of text and image analysis for flood event image recognition | |
Chae et al. | Identification of strategic fields for developing smart city in Busan using text mining | |
Riga et al. | Atmospheric environment and quality of life information extraction from twitter with the use of self-organizing maps | |
CN106407271B (en) | Intelligent customer service system and updating method of intelligent customer service knowledge base thereof | |
Lawu et al. | Social media data crowdsourcing as a new stream for environmental planning & monitoring: A review |
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 |