How can I determine trending stories from a collection of news data
I have a portal that fetches news from hundreds of resources around the web. How can I be able to use开发者_C百科 these data to determine trending stories ?
Any ideas would be highly appreciated.
Thanks.
- Get a good dictionary with all kinds of words and their forms.
- Split the words in a news story and collect statistics on those words.
- Group stories by their word statistics signature and closeness in date-time space.
精彩评论