A dataset is the source and the starting point of your analysis. It is the data from which you will extract insights. 

On the Symplur platform we have 4 types of datasets:

  1. Keywords. Keyword dataset types can consist of one or more words, for example "diabetes", "lung cancer", "non-small cell lung carcinoma". All usage of this keyword or phrase will be found in that dataset. Twitter sends Symplur 100% of all tweets that includes the phrase "lung cancer". 
  2. Hashtags. This is an easy dataset type. We get all tweets that contains the hashtag "#LCSM".
  3. Twitter Accounts. This dataset contains all tweets from a certain account and also all tweets mentioning that same Twitter account. In other words, this dataset type contains bi-direction conversations with an account. Example would be "@symplur". 
  4. URLs. This dataset type will contain all tweets that contain certain urls regardless of the text in the tweet. These urls can be specific url or it can be broad such as a top-level domain name. Examples: "www.nejm.org/doi/full/10.1056/NEJMoa18026371"

Did this answer your question?