The survey explores the world of data mining in terms of the chances, the expectations, the future trends, and of course its players. The survey contained 20 questions about:
- software usability
- quality and variety of tools
- speed and large data sets
1319 users completed the survey evaluating the major software companies for data mining. KNIME came out of this survey quite well. In general. KNIME performed above average, besides for documentation where it just received an average degree from its users.
Somehow, I expected that KNIME being open source would occupy the top part for the ranked list in terms of cost.
I also had high expectations in terms of ease of use. KNIME GUI is very intuitive to use and you do not need a months long course to learn to write a dedicated script.
The quality of the tool and the variatey of data manipulation nodes and data mining algorithms also does not need much of a discussion. Among all the tools I have used, KNIME contains the most popular data manipulation and data mining algorithms that I need to use.
What surprised me is that KNIME came on top of the list also in terms of speed and for handling large data. It is true that KNIME rarely gets stuck when dealing with millions of rows. It might become slow, sure. But if you are patient enough and your machine is powerful enough, eventually all data get processed. It is comforting however to discover that KNIME performs well in terms of speed and large data sets if compared with the largest also non-open source players in the field.
This is a very encouraging result for KNIME and its crew. It is also a very encouraging result for us users. Whyever we started using KNIME, we feel now that it has been a very well grounded choice.
The part of the survey evaluating the data mining siftware is available here.