Difference between retrieving tweets using the Streaming API and the REST API



I’m planning to do an application that retrieves tweets and then we will perform some data mining on those tweets.

Using the Streaming API I can choose to use public streams (that as per my understanding have a “neutral” or “public” point of view of data gathering), user streams (user oriented point of view data gathering) and site streams (that is something like a set of users point of view data gathering). Ok. Now regarding the REST API, that is used by GET search/tweets: what is the point of view of the tweets that I’ll get from this type of search? I can see that an authentication with a certain twitter account is required, and because of that I am assuming that all resulting tweets from the GET search/tweets are gathered from a user-oriented point of view (similar to user streams) and NOT from a “neutral” point of view perspective, that is the most desirable choice to a data mining application. If I use a user-oriented point of view my searches will be affected by the user that I’m using and it’s characteristics like who it is following and etc, and this is not what I want.

I need a confirmation regarding my assumptions and please correct me if I’m wrong.

Thanks and have a nice day,