We just raised a $30M Series A: Read our story

Badges

115 Points
3 Years

User Activity

About 1 month ago
I have not used it - sorry
4 months ago
I suspect that I cannot answer this. I have used Knime and RapidMiner with data sets that have had up to about 80,000 rows and 1,500 columns and both have performed well. However, I doubt whether the questioner would classify my usage as "large amounts of data". If my usage…
4 months ago
I have found both Knime and RapidMiner easy to use and powerful. That applies to producing numerical output, graphical representations, and PMML files. The latter helps if one wishes to produce diagnostic or recommender systems automatically, with no human intervention…
Almost 2 years ago
I used IBM Modeler several years ago and found it to be effective, but expensive. Fortunately, it was for a commercially funded contract. For KNIME I have only used it for experimental purposes and found it rather cumbersome but powerful. It is also more cost-effective. I…
Over 3 years ago
Safeguarding naive users against erroneous reporting from not knowing the statistical assumptions underlying a given technique i.e. I am agreeing with Nicholas Kogan. After that, the order of importance of features depends on the use, and on who the user will be. The…

About me

Application of scientific methods to social and medical problems - statistical analyses, expert systems, machine learning. I have developed a number of novel methods of making social research replicable.

Specialties: Data analysis, SPSS, Answer Tree, STAS, various rule-based and case-based expert system packages (they change over time as the technology develops)