In order to evaluate our community detection on an underlying application (extrinsic evaluation), we choose news recommendation. For this evaluation, we need to build a gold standard of news articles and the users who mentioned the url of the news articles in their tweets, meaning that they are interested in the topics of the news article (polarity does not matter for our sake).
So, we need to crawl the actual news article from mentioned urls in tweets to extract their topics.
Hi Soroush,
We can manage the tasks on your codeline from the todo list of your project and make them an issue like this.
Hope you find it useful.
In order to evaluate our community detection on an underlying application (extrinsic evaluation), we choose news recommendation. For this evaluation, we need to build a gold standard of news articles and the users who mentioned the url of the news articles in their tweets, meaning that they are interested in the topics of the news article (polarity does not matter for our sake).
So, we need to crawl the actual news article from mentioned urls in tweets to extract their topics.
Hi Soroush, We can manage the tasks on your codeline from the todo list of your project and make them an issue like this. Hope you find it useful.
Best, hf