Instead of looking at news coverage to detect trends, shouldn't I look instead at sales data or something? Maybe -- that's one way to measure popularity. But maybe popularity should be measured based on how talked about it is, in which case news coverage works fine. Then again, people may talk more about stuff that sells like hotcakes, and it doesn't matter.
In this graph we detect three new trends in women's clothing:
Skinny jeans and ballet flats just recently exploded in popularity, and the news coverage reflects that. I looked for sales data but could only find an industry report that would cost too much to buy. It surely shows something like the above. As for boy shorts, I reviewed a few news articles on the decline of slutty styles, and the sales data show that boy shorts became popular in 2004 and remain so today. Sure enough, that's just what the news coverage tells us.
So, I'm not too worried about using news coverage as opposed to sales data.