Separating Junk Data In Big Data

There is no simple answer for these questions, because only when questions are asked a data-analyst looks into the data to find answers. Then it is information that is important. The data-analyst must be good in looking into the Big Data for information and separating it for the junk data.

Big Data

Big Data is every data collected which has no meaning yet. The difference between Big Data and other data collection is the Big Data has the following features:

• Volume --> Everything is registered, kept and stored

• Velocity --> It is the information age, which means everything goes fast and is kept because there is no time to think about what to keep or not

• Variety --> The data is kept, all the data. For example the time on which a survey is filled in, which most of the time has no meaning when analyzing the survey data.

Big Data Source

There are now 3 kinds of Big Data sources. In the future there could be more as society goes deeper into the information age. The 3 sources:

• Collected by people --> Everything people collect digital. I have a collection of tweets online and on my computer more digitally. Not all that I have collected will be used by me in the future, therefor it is Big Data

• Registration of process --> As told before when I fill in a survey the time when I fill in the survey is registered, this is a process which is not necessary for the research for which the survey is meant. The registered time is most of the time unimportant, but when something goes wrong with the survey it can be used to trace the survey

• Collected by machines automaticly --> When I use electricity at home this data is collected. If I use Toon (Eneco) it would registered which place in my home and which appliance I would have used. This is done automaticly I only have to use Toon. This is important if I want to know which appliance uses the most electricity.