Cluster analysis for data
Budget: ₹1,500 – ₹3,000 INR
Hey, I am Jareer.
Uploaded file is an excel file and it contains 6 columns(time, minimum, mean, maximum, number of observations and day). Minimum, mean and maximum columns denote the minimum, mean and maximum value of travel time for each 5 minute interval for a particular date. Each day is divided into 5 minute intervals from 6 am to 11 pm. There are 61 such days.
The work:
Cluster analysis of data, preferably in Python, Matlab.
For each time interval, cluster each of the minimum, mean and maximum columns individually.
Eg: For 6 am, first cluster all the rows in the minimum column. Separately, cluster mean and maximum columns.
Decide on a fixed optimal number of clusters for all three columns(minimum, mean and maximum).
Each cluster should be labelled with a different color.
For each time interval, find the centroidal value for each cluster for each of minimum, mean and maximum columns.
Eg: For 6 am, if there are 3 clusters for the minimum column, find the centroidal value for all 3 clusters. Similarly proceed for other time intervals and maximum/mean columns.
Write these centroidal values for each column and time interval in an excel file.
Store the cluster plots for each column and time period in a word file.
Do the cluster analysis for each 5 minute interval from 6 am to 1 pm. Leave the remaining time intervals.
Uploaded file is an excel file and it contains 6 columns(time, minimum, mean, maximum, number of observations and day). Minimum, mean and maximum columns denote the minimum, mean and maximum value of travel time for each 5 minute interval for a particular date. Each day is divided into 5 minute intervals from 6 am to 11 pm. There are 61 such days.
The work:
Cluster analysis of data, preferably in Python, Matlab.
For each time interval, cluster each of the minimum, mean and maximum columns individually.
Eg: For 6 am, first cluster all the rows in the minimum column. Separately, cluster mean and maximum columns.
Decide on a fixed optimal number of clusters for all three columns(minimum, mean and maximum).
Each cluster should be labelled with a different color.
For each time interval, find the centroidal value for each cluster for each of minimum, mean and maximum columns.
Eg: For 6 am, if there are 3 clusters for the minimum column, find the centroidal value for all 3 clusters. Similarly proceed for other time intervals and maximum/mean columns.
Write these centroidal values for each column and time interval in an excel file.
Store the cluster plots for each column and time period in a word file.
Do the cluster analysis for each 5 minute interval from 6 am to 1 pm. Leave the remaining time intervals.