I Like, I Love, I Want Some More of It.
In information retrieval a few strong words. They really descriptive and get right to the point of what people are looking for. Other words have little value does not exist. The reason the concept of stop words appear is that you really can not say much about the documents that include words such as, an, and, that, etc.. The other side of stop words are words that have a high discrimination values. Recently I was looking to see if there are FedEx office in the town where my mother lived, and although no one, Google still returned a few pages (home pages and page store locator) from FedEx.com web site in search results. That's great search results, and Google is smart to place more weight on core concepts in a search word (FedEx), while placing less weight on the location. Words that have a low discrimination value may have a higher discriminatory value when combined with neighboring words. Heat and dogs may have different meanings when they are adjacent. As described in thi