search

sjkp / heatmap

The PowerBI Heatmap Visual enables users to draw a heatmap overlay from a X, Y coordinate set on to an existing image. The user specify the image, and provide a data set of X, Y coordinates and optionally an intensity for each data point. The radius and the bluriness of the heatmap bubbles can be customized as well as the max value for the intensity.
20 stars 7 forks source link

Can we do weighted analysis heat map visual? #7

Open EshwarKV opened 6 years ago

EshwarKV commented 6 years ago

Hi sjkp,

Thanks for the awesome heat map custom visual.

Can we do weighted analysis heat map visual by including weight variable? Also I want to know for what purpose the Intensity option is used in the Fields Tab.

It would be appreciated if you provide the comments ASAP.

Thank you, Eshwar

sjkp commented 6 years ago

Can you explain to me what you mean by weight variable? I'm not sure I understand.

If you dont want all x,y coordinates to mean the same. E.g. you have click postions on a web page and for each x,y coord you record the number of clicks then you can put that as the intensity.

EshwarKV commented 6 years ago

Thanks for your response. I got the what Intensity stands for from your explanation.

My concern is similar to this, I guess!!! but let's put on discussion.

Following is the example data:

Resp ID | X | Y | Weight 1 | 437 | 333 | 1.4 2 | 94 | 535 | 1.2 3 | 120 | 372 | 0.8 4 | 269 | 671 | 0.9 5 | 115 | 580 | 0.7

Mean | 207.0 | 498.2 |   Wtd. Mean | 228.0 | 482.5 |  

Here I have mentioned the weighed analysis what we do regularly in summary tables and I want to do the similar thing in heat map visualization.

Here the Mean for X (or Y) are calculated as total of X's/total no. of respondents whereas Wtd. Mean was calculated as sum of product of (X and it's respective weight value) / sum of weight. Here the weights are given to respondents in such a way that the sum of weights must be equal to the number of respondents.

Can we do this weighted visualization in heatmap as we do weighted mean, std and etc... in weighted analysis? Many thanks in advance.