Matlab: organize a great number of data based on month, day and hourly intervals
1 view (last 30 days)
Show older comments
Hi! I have a problem and I'm not able to find a smart solution. I explain it in short: I have a great number of user's sequence of location records and I have clustered them with cluster alghoritm optics. After find the cluster, I find the sequences of location records in every cluster. They are included in SetOfDataset (attached) end that is a struct 1x50, where 50 is the number of the cluster. Every sequence of locations has = [userId year month day hour minutes seconds latitude longitude locationID]
I want to divide this sequences based on month, day and, after identified the days, I want to divide them in intervals of 2 hours.
I have try to use nested struct but the code is terrible: slow, not clear the results... I want to do something of easy but I can't find smart ideas: can you help me?
2 Comments
Accepted Answer
Guillaume
on 16 Jun 2016
It's not clear what final output you're looking for. Something like this:?
for setidx = 1 : numel(SetOfDataset)
points = SetOfDataset(setidx).points;
%divide date/hour in groups of two hours:
[~, ~, dateid] = unique([points(:, 2:4), floor(points(:, 5)/2)], 'rows', 'stable'); %'stable' optional
%use that to split the point array into cell arrays
SetOfDataset(setidx).splitpoints = arrayfun(@(did) points(did == dateid, :), [1:max(dateid)]', 'UniformOutput', false);
end
3 Comments
Guillaume
on 16 Jun 2016
"I want also to divide the hours associated to every day in intervals of 2 hours"
It is doing this already. That's the purpose of the floor(points(:, 5)/2)
More Answers (0)
See Also
Categories
Find more on Dates and Time in Help Center and File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!