CN113626668A

CN113626668A - Map-oriented news multi-scale visualization method

Info

Publication number: CN113626668A
Application number: CN202110747359.5A
Authority: CN
Inventors: 罗运; 胡宏伟; 余思佳; 罗彩玉; 蔡忠亮
Original assignee: Wuhan University WHU
Current assignee: Wuhan University WHU
Priority date: 2021-07-02
Filing date: 2021-07-02
Publication date: 2021-11-09
Anticipated expiration: 2041-07-02

Abstract

The invention relates to a multi-scale visualization method of news, which is used for realizing multi-scale expression of network news facing to a map and comprises the following steps: 1) crawling different news events in a certain time period, acquiring news text content, a news title and news release time, and storing news sources; 2) geocoding the crawled news to acquire news places and corresponding longitude and latitude coordinates; 3) carrying out scale division on news according to place names related in the news; 4) carrying out span calculation on news according to the longitude and latitude of the place involved in the news; 5) and constructing a map visualization model according to the dimensions and the spans of the news, so that the news with corresponding dimensions and spans are displayed according to different map zooming dimensions. According to the invention, the news scale and the map scale are combined, so that the multi-scale presentation of news data is realized, the browsing dimension of news information is increased, the browsing and reading of news are more convenient and efficient, and the multi-scale news visualization technology under the space-time dimension is realized.

Description

Map-oriented news multi-scale visualization method

Technical Field

The invention relates to geographic information science and geographic information visual expression, in particular to a news multi-scale visual method.

Background

In the big data era, news information is used as an effective way for acquiring information, and the problems of information redundancy, unclear classification, non-visual expression and the like exist. The map is used as an important information carrier, has a special language system, and has the characteristics of intuition and multiple dimensions in the aspect of information display. News information is aggregated by taking the geographic dimension as a basis and the time dimension as a variable, so that news space-time map interaction on the geographic dimension can be realized.

However, the news map expression at the present stage lacks the definition of the scale, and cannot embody the scale characteristics of the news. This results in the inability to combine the map scale with the news scale to highlight the scale meaning of the news when the news is visually presented on the map. Therefore, a method for defining news dimension and news span for performing multi-dimension expression of news on a map is lacked at present.

Disclosure of Invention

The invention aims to overcome the defects of the prior art and provide a news visualization method which is visual and vivid and embodies different scales of news, so that the multi-scale expression of the news on a map is realized, and the multi-scale news visualization technology under the space-time dimension is realized.

The technical scheme provided by the invention is a news multi-scale visualization method, which comprises the following steps:

step S10, crawling different news events in a certain time period, acquiring news text content, titles and release time, and storing metadata such as news sources;

step S20, geocoding the crawled news to obtain news places and corresponding longitude and latitude coordinates;

step S30, according to the place name involved in the news, carry on the scale division to the news;

step S40, according to the longitude and latitude of the place involved in the news, span calculation is carried out on the news;

and step S50, constructing a map visualization model according to the dimensions and the spans of the news, so that the news with the corresponding dimensions and spans are displayed according to different map zooming dimensions.

Further, the step S10 of acquiring the content of the news text specifically includes:

step S101, first, by reading the HTML code of the web page, the web page is parsed into a DOM tree, each HTML tag is a node, and all texts are leaf nodes in the DOM tree. Traversing each node in the DOM tree, and calculating the total number of character strings of all text leaf nodes in the node by the total number of child nodes contained in the node to obtain the text density of the node. And screening the node with the highest text density in the DOM tree to obtain a news text DOM node. Further, the news text is obtained by acquiring the text content of the text leaf node in the DOM node.

In step S102, it is determined whether the given content is a link or a text. For text, it is directly taken as news text for the subsequent steps.

And S103, judging a news website for the link, and if the link website is a news portal website except for WeChat and microblog, processing the link website in the step S201 to extract a news text. And if the link website is a Sino microblog or a WeChat chapter, using CSS and XPath selectors for the link website, acquiring class, id and data attributes of a DOM node containing the news text according to DOM structures of the microblog and WeChat pages, and acquiring the news text by using CSS and XPath rules corresponding to the attribute.

Further, step S20 specifically includes:

step S201, a place name database of provinces, cities and prefectures in China is constructed, and the place names are classified according to the grades of the provinces, the direct prefecture cities, the non-direct prefecture cities and the prefectures;

step S202, comparing news text contents by using the constructed place name database, extracting province names, city names and district names which are stored in the place name database, and forming a province name list, a city name list and a district name list;

step S203, adding corresponding district names to the district name list of the district names identified in the news text; adding corresponding province names to a province name list for city names identified in a news text;

step S204, geocoding news places related to news to obtain longitude and latitude coordinates of the news places.

Further, the specific implementation of step S204 includes:

in step S2041, a list of geocoding ranges is determined. If the number of entries in the district name alternative list is greater than 1, the geocoding range list is the district name alternative list, otherwise, whether the number of entries in the city name alternative list is greater than 1 is judged, if yes, the geocoding range list is the city name alternative list, otherwise, whether the number of entries in the province name alternative list is greater than 1 is judged, if yes, the geocoding range list is the province name alternative list, and if not, the geocoding range list is nationwide.

Step S2042, traversing the place name alternative list, and if the geocoding range is not nationwide, performing local geocoding service calling on the current place name by taking the geocoding range as city limit according to the entries in the geocoding range list. The local geocoding service returns geographic latitude and longitude coordinates under a BD09 coordinate system of the query keyword according to the specified query keyword, the search geographic range limiting parameter, the search city limiting parameter and the search category limiting parameter by searching a Baidu map national POI point-of-interest database, and simultaneously returns the comprehension score and the credibility score of the coding result. The comprehension degree score can be used for judging whether the incoming query key words exist in a POI interest point database and the similarity of the incoming query key words and data in the POI interest point database, and the higher comprehension degree indicates that the name format of the place name is more correct and the possibility of finding corresponding coordinates in the database is higher; the credibility score can be used for judging the place name range related to the inquired place name, and the higher the credibility score is, the more specific the position related to the place name is. In order to ensure the correctness of the place name resolution, an understanding score threshold value is set to be 70, and a credibility score threshold value is set to be 20, so that wrong place name resolution results are filtered. And judging the result of the hundredth geocoding service, and if the understanding score of the result of the geocoding service is greater than 70 and the credibility score is greater than 20, storing the encoding result into an encoding result candidate list.

And if the geographic coding range is national, calling national geographic coding service for the place name. The national geocoding service returns geographic longitude and latitude coordinates under a Mars coordinate system according to the specified query key word parameters by searching a national POI interest point database of the Gaode map. The national geocoding service has higher accuracy for coding national place names without designating the province scope of a city, and the geographical coding result is returned to indicate that the place names are correct and unique and accurate longitude and latitude coordinates under a mars coordinate system can be returned. And if the returned result is not null, storing the encoding result into an encoding result candidate list.

Step S3043, traversing the candidate list of encoding results, for the same-name place names, selecting an entry with a higher understanding score and a higher credibility score, and deleting other entries.

Further, step S30 specifically includes:

step S301, defining provincial scale of news. Judging the number of province names related to the news text, if the number of province names is larger than one, determining that the news scale is a multi-province, if the number of province names is equal to one, determining that the news scale is a single province, and if the number of province names is 0, defining that the news scale is a national scale;

step S302, a city-level scale of news is defined. Judging the number of city names related to the news text, if the number is larger than one, confirming that the news scale is a multi-city, if the number is equal to one, determining that the news scale is a single city, and if the number is 0, defining the city scale of the news as empty;

step S303, a prefecture-level scale of news is defined. Judging the number of names of districts and counties related to the news text, if the number of names of the districts and the counties is larger than one, confirming that the news scale is a multi-county, if the number of names of the districts and the counties is equal to one, confirming that the news scale is a single county, and if the news scale is 0, defining the county-level scale of the news as null;

and S304, integrating province scale, city scale and district scale of news to form province-city-district scale.

Further, step S40 specifically includes:

step S401, if the number of news places is unique, defining the span to be 0; if the number of the news places is not unique, aggregating the single-point elements of the places corresponding to the news to form multi-point elements;

and S402, calculating the diameter of the minimum circumscribed circle of the multipoint elements by using the minimum circumscribed circle function provided by the PostGIS, wherein the diameter is used as the span of news.

Further, step S50 specifically includes:

step S501, the scale of news is matched according to the zoom scale of the map, so that news with different scales is displayed under different zoom scales of the map;

step S502, according to the span of news, news places are displayed on the map in different point sizes. For a point with news span of 0, the diameter of the display point is the largest; as the news span increases, the size of the place display diameter corresponding to the news is continuously reduced.

Compared with the prior art, the invention has the following advantages:

1. by carrying out different news content extraction methods on different news sources, the news content extraction of each news portal website is effectively compatible, and the application range of news geocoding is wider.

2. A place name database is constructed according to province, city and district names in the country, and news contents are matched, so that the range of calling geocoding services is limited, and a geocoding result is more accurate;

3. the scale of news and the scale of a map are combined, multi-scale presentation of news data is achieved, browsing dimensionality of news information is increased, and browsing and reading of news are more convenient and efficient.

4. The news with different scales is displayed on the interactive map, so that the interactivity of the news is improved, and the browsing of the news is more visual, vivid and lively.

Drawings

FIG. 1 is a flow chart of the method of the present invention.

Detailed Description

For a further understanding of the invention, reference will now be made to the preferred embodiments of the invention by way of example, and it is to be understood that the description is intended to further illustrate features and advantages of the invention, and not to limit the scope of the claims.

The invention provides a map-oriented news multi-scale visualization method, the flow of which is shown in figure 1, and the method comprises the following steps:

step S10, crawling different news events in a certain time period, acquiring news text content, titles and release time, and storing metadata such as news sources; for the example, 61800 news events in 2019, month 10, month 2 to 2020, month 5 are crawled, and the content of news, such as the title, the text, the release event and the like, as well as the element data of an author, a source and the like are obtained and stored.

Step S20, geocoding the crawled news to obtain news places and corresponding longitude and latitude coordinates; for this example, a total of 335727 longitude and latitude coordinates were obtained.

In an embodiment, the step S10 of acquiring the content of the news text specifically includes:

For this example, for a given news linkhttps://weibo.cn/comment/JcJrUrciWAnd extracting news text.

In an exemplary embodiment, step S20 specifically includes:

For this example, for a given news linkhttps://weibo.cn/comment/JcJrUrciWThe province name list is Liaoning province, the city name list is Dalian city, and the district name list is Ganjin subregion and West Bao subregion. The obtained news latitude and longitude coordinates are as follows:

"incense burner reef street worker village

"longitude":121.59200254824019,

"latitude":38.933387294420534

},

'Dalian bay street', a retaining pocket

"longitude":121.7079030595669,

"latitude":39.0238804916907

},

"incense burner reef street

"longitude":121.59951939123108,

"latitude":38.93593059882904

},

"Daliankayo world seafood GmbH": great face

"longitude":121.96799695008666,

"latitude":39.40102888217002

}

In an exemplary embodiment, step S30 specifically includes:

and S304, integrating province scale, city scale and district scale of news to form province-city-district scale. For this example, for a given news link https:// weibo. cn/comment/jcjrrurciw, the province scale is a single province scale, the city scale is a single city scale, and the prefecture-level scale is a multi-prefecture scale.

In an exemplary embodiment, step S40 specifically includes:

and S402, calculating the diameter of the minimum circumscribed circle of the multipoint elements by using the minimum circumscribed circle function provided by the PostGIS, wherein the diameter is used as the span of news. For this example, for a given news linkhttps://weibo.cn/comment/ JcJrUrciWAnd the news span is 0.317466267006714.

In an exemplary embodiment, step S50 specifically includes:

step S501, the scale of news is matched according to the zoom scale of the map, so that the news with different scales is displayed under different zoom scales of the map, and the division mode is shown as table 1;

TABLE 1

Step S502, according to the span of news, news places are displayed on the map in different point sizes. For a point with news span of 0, the diameter of the display point is the largest; as the news span increases, the size of the place display diameter corresponding to the news is continuously reduced. For this example, for a given news linkhttps://weibo.cn/comment/JcJrUrciWThe news is displayed on the map when the zoom level of the network map tile specification is greater than 9, and each point of the news is displayed in size of Exp (-0.317466267006714) units, that is, 0.7279912 units.

The above description of the embodiments is only intended to facilitate the understanding of the method of the invention and its core idea. It should be noted that, for those skilled in the art, it is possible to make various improvements and modifications to the present invention without departing from the principle of the present invention, and those improvements and modifications also fall within the scope of the claims of the present invention.

Claims

1. A map-oriented news multi-scale visualization method is used for realizing multi-scale expression of news on a map, and comprises the following steps:

step S10, crawling different news events in a certain time period, acquiring news text content, titles and release time, and storing news sources;

2. The map-oriented news multiscale visualization method as recited in claim 1, wherein: the obtaining of the news text content in step S10 specifically includes,

step S101, firstly, a webpage is analyzed into a DOM tree by reading a webpage HTML code, each HTML label is a node, and all texts are leaf nodes in the DOM tree; traversing each node in the DOM tree, and calculating the total number of character strings of all text leaf nodes in the node by the total number of child nodes contained in the node to obtain the text density of the node; and screening the node with the highest text density in the DOM tree to obtain a news text DOM node. Further, obtaining a news text by obtaining the text content of a text leaf node in the DOM node;

step S102, judging whether the given content is a link or a text; directly taking the text as a news text to carry out subsequent steps;

s103, judging a news website for the link, and if the link website is a news portal website except for WeChat and microblog, processing the link website in the step S201 to extract a news text; and if the link website is a Sino microblog or a WeChat chapter, using CSS and XPath selectors for the link website, acquiring class, id and data attributes of a DOM node containing the news text according to DOM structures of the microblog and WeChat pages, and acquiring the news text by using CSS and XPath rules corresponding to the attribute.

3. The map-oriented news multiscale visualization method as recited in claim 1, wherein: the step S20 specifically includes the steps of,

4. The map-oriented news multiscale visualization method as recited in claim 1, wherein: specific implementations of the step S204 include that,

step S2401, determining a geocoding range list; if the number of entries in the district name alternative list is greater than 1, the geocoding range list is the district name alternative list, otherwise, whether the number of entries in the city name alternative list is greater than 1 is judged, if yes, the geocoding range list is the city name alternative list, otherwise, whether the number of entries in the province name alternative list is greater than 1 is judged, if yes, the geocoding range list is the province name alternative list, otherwise, the geocoding range list is nationwide;

step S2402, traversing a place name alternative list, and if the geographic coding range is not national, calling hundredth geographic coding service for the current place name by taking the geographic coding range as city limitation according to entries in the geographic coding range list, wherein the hundredth geographic coding service returns geographic longitude and latitude coordinates under a BD09 coordinate system of a query keyword according to a specified query keyword, a search geographic range limiting parameter, a search city limiting parameter and a search category limiting parameter by searching a hundredth map national POI interest point database, and simultaneously returns an comprehension score and a credibility score of the coding result; the comprehension degree score can be used for judging whether the incoming query key words exist in a POI interest point database and the similarity of the incoming query key words and data in the POI interest point database, and the higher comprehension degree indicates that the name format of the place name is more correct and the possibility of finding corresponding coordinates in the database is higher; the credibility score can be used for judging the place name range related to the inquired place name, and the higher the credibility score is, the more specific the position related to the place name is; if the understanding score of the coding service result is greater than P1 and the credibility score is greater than P2, storing the coding result into a coding result candidate list;

if the geographic coding range is national, national geographic coding service calling is carried out on the place names, and the national geographic coding service returns geographic longitude and latitude coordinates under a Mars coordinate system according to specified query keyword parameters by searching a national POI interest point database of a Gagde map;

step S2403, traversing the encoding result candidate list, selecting an item with higher understanding score and credibility score for the same place name, and deleting other items.

5. The map-oriented news multiscale visualization method as recited in claim 4, wherein: the understanding score threshold is set to 70, and the credibility score threshold is set to 20, namely P1 is set to 70 and P2 is set to 20.

6. The map-oriented news multiscale visualization method as recited in claim 1, wherein: the step S30 specifically includes the steps of,

step S301, defining province scale of news; judging the number of province names related to the news text, if the number of province names is larger than one, determining that the news scale is a multi-province, if the number of province names is equal to one, determining that the news scale is a single province, and if the number of province names is 0, defining that the news scale is a national scale;

step S302, defining a city-level scale of news; judging the number of city names related to the news text, if the number is larger than one, confirming that the news scale is a multi-city, if the number is equal to one, determining that the news scale is a single city, and if the number is 0, defining the city scale of the news as empty;

step S303, defining a district-level scale of news; judging the number of names of districts and counties related to the news text, if the number of names of the districts and the counties is larger than one, confirming that the news scale is a multi-county, if the number of names of the districts and the counties is equal to one, confirming that the news scale is a single county, and if the news scale is 0, defining the county-level scale of the news as null;

7. The map-oriented news multiscale visualization method as recited in claim 1, wherein: the step S40 specifically includes the steps of,

8. The map-oriented news multiscale visualization method as recited in claim 1, wherein: the step S50 specifically includes the steps of,

step S502, according to the news span, displaying news places on a map in different point sizes, and for the point with the news span of 0, displaying the point with the largest diameter; as the news span increases, the size of the place display diameter corresponding to the news is continuously reduced.