- This topic has 3 replies, 2 voices, and was last updated 1 year ago by
.
- Posts
Hi, I was wondering under which of the V’s it would be best to mention an issue like spelling mistakes, grammatical errors in non structured qualitative data (eg social media) etc.
On the one hand I can see that Veracity could be the answer. As a typo could deem that piece of information as unreliable.
However, I think I prefer to put it under Variety. If someone makes a typo, then the ‘better’ algorithm would be one that recognises what it should say and still includes it.
Would be grateful to hear your thoughts.
Thanks
I would go for veracity.
For example, if you spell a name incorrectly then you might identify in incorrect person or company. In addition, if you were did a search to look for all customers called ‘Smith’ but had entered one of them as ‘Smitg’ you would miss that person.
Variety means that the data is capable of holding a wide variety of types of data (Smitg as well as Smith, perhaps with pictures of these people). With many types of data it will be difficult to identify mistakes. For example, there might be some people called ‘Smitg’.
- The topic ‘Big data and spelling mistakes/typos’ is closed to new replies.