New in Wolfram Language: WikipediaData feature for integrating with Wikipedia and processing its data
Since the creation of the Wolfram | Alpha service , Wikipedia has occupied a special place in its development. We usually use it not as the primary source of data, but rather as the most important resource for improving the recognition of natural language. In particular, for the extraction of data on how people describe certain things in the spoken / official style.
For many years we have been developing various tools for analyzing and extracting information from Wikipedia, but now we are adding an “integration service” with Wikipedia, which will be available in the new version of the Wolfram Language (Mathematica 10.1, coming out very soon). Now, embedding Wikipedia content into workflows inside the Wolfram Language has become much easier.
Of course, you can simply take the text from the Wikipedia article and pass it on to the new Wolfram Language functions for word processing and visualization:
')
If you can not specify the exact title of the article, you can search by name or content:
You can even use Wolfram Language objects directly in WikipediaData to, say, find similar articles in some other languages ​​that are on Wikipedia.
One of my favorite features allows you to explore links to articles within a particular article or category. And the main charm is that the data can be displayed either using a simple list, or use the various functions and rules of the Wolfram Language for visualization in the form of, say, a graph. In fact, with just a few lines of code, you can create a beautiful and interesting visualization of the links between any set of Wikipedia articles:
This is just the tip of the iceberg, and this feature can do many other useful things. Get a free subscription to Wolfram Programming Cloud to see what you can do with WikipediaData after the release of the new version of Wolfram Language, and do not miss the integration releases with other services that will be released next year.
For many years we have been developing various tools for analyzing and extracting information from Wikipedia, but now we are adding an “integration service” with Wikipedia, which will be available in the new version of the Wolfram Language (Mathematica 10.1, coming out very soon). Now, embedding Wikipedia content into workflows inside the Wolfram Language has become much easier.
Of course, you can simply take the text from the Wikipedia article and pass it on to the new Wolfram Language functions for word processing and visualization:
')
If you can not specify the exact title of the article, you can search by name or content:
You can even use Wolfram Language objects directly in WikipediaData to, say, find similar articles in some other languages ​​that are on Wikipedia.
One of my favorite features allows you to explore links to articles within a particular article or category. And the main charm is that the data can be displayed either using a simple list, or use the various functions and rules of the Wolfram Language for visualization in the form of, say, a graph. In fact, with just a few lines of code, you can create a beautiful and interesting visualization of the links between any set of Wikipedia articles:
This is just the tip of the iceberg, and this feature can do many other useful things. Get a free subscription to Wolfram Programming Cloud to see what you can do with WikipediaData after the release of the new version of Wolfram Language, and do not miss the integration releases with other services that will be released next year.
Source: https://habr.com/ru/post/254345/
All Articles