Reader Comments

Post a new comment on this article

Making the raw data publicly available would be great.

Posted by rvencio on 22 Apr 2008 at 18:47 GMT

This paper is very interesting... congratulations to the authors. I believe that, in the spirit of PLOS NTD, it would be very good if the authors could release the raw data as one of the Supporting Information files (given the appropriate "anonymization").

I understand and agree with the author's claim: "Decision algorithms are also easy to interpret, use and validate using common statistical techniques. Importantly, it provides a means to identify parameters that would be significantly associated with disease when analysed in sub-groups but not in the total study population." However, one should be open to the possibility of more complex methods/algorithms providing even better performance. Therefore, I would argue that making the raw data available to the whole community would multiply the data's value and worth some extra citations to this nice paper.

Recently in Brazil we experienced a regrettable dengue outbreak and our medical/scientific community could benefit a lot from this data obtained by Tanner et al.

I hope the authors understand and agree with this position and can release the raw data for further experimentation with other machine learning techniques that, although not as elegant as the decision trees used here, could potentially yield slightly better results.


RE: Making the raw data publicly available would be great.

engeong replied to rvencio on 28 Apr 2008 at 03:56 GMT

Thank you for your interest in our work. We would be delighted to make our raw data from Singapore available to you and anyone who might be interested. Please contact corresponding author for more details.