Timeline for Variables involved in kNNdistplot (dbscan package) in R
Current License: CC BY-SA 3.0
9 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Feb 25, 2016 at 11:22 | comment | added | tushaR | Let us continue this discussion in chat. | |
| Feb 25, 2016 at 11:18 | comment | added | Has QUIT--Anony-Mousse | You need to write down your task mathematically. What is the objective function that you want to optimize/compute? | |
| Feb 25, 2016 at 11:15 | comment | added | Has QUIT--Anony-Mousse | So you have 4 time series that you want to correlate. I don't see clustering there. Clustering would make sense if you have thousand servers and no idea where they are located, and expect them to suffer from data center problems. | |
| Feb 25, 2016 at 11:11 | comment | added | tushaR | Okay. My input is 4 time series of the same feature corresponding to 4 servers. The length of each time series is 2000. The objective is to determine that time-series of which servers fall in the same clusters using these series. Can we attain the objective using DBSCAN ? | |
| Feb 25, 2016 at 9:56 | comment | added | Has QUIT--Anony-Mousse | To do what? I don't see how you could use DBSCAN to solve anything on your data. Plus, that article is in a low reputation indian spam journal... I would not bother to even read it. | |
| Feb 25, 2016 at 9:39 | comment | added | tushaR | I am trying to calculate the parameters based on the algorithm provided in this paper: ijiset.com/v1s4/IJISET_V1_I4_48.pdf | |
| Feb 25, 2016 at 9:26 | comment | added | Has QUIT--Anony-Mousse | I think you first need to be clear about what question you are asking. I don't even understand what parameter you are trying to estimate, nor how you mean to use it. | |
| Feb 25, 2016 at 6:32 | comment | added | tushaR | So my approach is wrong. I won't argue on this as I was experimenting. Now I know where I am going wrong. Do you think it would help if I use the entire 8000 datapoints as one data set and calculate my parameters based on the distance of every point from the remaining 7999 points? | |
| Feb 24, 2016 at 20:35 | history | answered | Has QUIT--Anony-Mousse | CC BY-SA 3.0 |