Timeline for Categorization/Segmentation techniques
Current License: CC BY-SA 3.0
10 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 10, 2011 at 3:34 | vote | accept | Colin K | ||
| May 9, 2011 at 22:51 | answer | added | SheldonCooper | timeline score: 5 | |
| May 9, 2011 at 21:22 | comment | added | Colin K | @deps_stats: The only one of those I am familiar with is k-means, which I have used in the past for auto-thresholding images. In this case, looking at scatter plots of my data has convinced me that the data does not cluster into well defined populations. At least, not in a way that well reflects the human-defined classification. I've already left my office for the day, but I think scatter plots would be acceptable for me to share, soI'll make some edits tomorrow. | |
| May 9, 2011 at 21:13 | comment | added | deps_stats | @Colin Have you tried any clustering analysis by now? (k-means, hierarchical clustering, partitioning around medoids...) | |
| May 9, 2011 at 18:38 | comment | added | Colin K | @deps_stats: I wish my data were more like signals. I'm much more comfortable with signal processing than with this sort of statistical modeling/analysis. | |
| May 9, 2011 at 18:34 | comment | added | deps_stats | @Colin No problem. I'm currently working on market segmentation for a pharmacy wholesaler. The problem is that their data look a lot like signals. I see a lot of similarities between my problem and your problem. I'll try to post some research that might help you. | |
| May 9, 2011 at 16:56 | comment | added | Colin K | @deps_stats: Thank you for the encouragement, I'm glad to see that people think I've posed my question at least somewhat decently. Unfortunately, my data is probably not something that should be on a public website. It isn't exactly classified, but its definitely not for public consumption. I will try to generate some artificial data that will serve as a good example for the purpose of this question, but I may not have time to do that for a day or so. | |
| May 9, 2011 at 15:03 | comment | added | deps_stats | @Colin It would be nice to have a look at the pictures. You've explained your problem with detail, it would be a nice complement. :) | |
| May 7, 2011 at 6:11 | history | tweeted | twitter.com/#!/StackStats/status/66746954952089600 | ||
| May 6, 2011 at 19:10 | history | asked | Colin K | CC BY-SA 3.0 |