Solved – How to visualize tags

data visualization

I have a data set which contains a hundred or so user tags – that is, binary variables marking whether a user has a certain property or has shown a certain behaviour. (data from a web site.)

user, high_usage, tablet, likes_cats, night_owl
A,    1,          0,      1,          0
B,    0,          1,      0,          1
C,    1,          0,      0,          0

How do I visualize this data? Pairwise or all together.

I'd like to create a visualization that allows exploration of this data to find groups of behavior.. Any ideas?

Best Answer

Given that what you have looks like an adjacency matrix how about using something like Gephi to visualise your data as a network. You could think of the tags as nodes and users as something which connets one or more nodes.

Related Solutions

Solved – Visualizing multiple “histograms” (bar-charts)

As you have found out there are no easy answers to your question!

I presume that you interested in finding strange or different book stores? If this is the case then you could try things like PCA (see the wikipedia cluster analysis page for more details).

To give you an idea, consider this example. You have 26 bookshops (with names A, B,..Z). All bookshops are similar, except:

Shop Z sells only a few History books.
Shops O-Y sell more romance books than average.

A principal components plot highlights these shops for further investigation.

Here's some sample R code:

> d = data.frame(Romance = rpois(26, 50), Horror = rpois(26, 100), 
               Science = rpois(26, 75), History = rpois(26, 125))
> rownames(d) = LETTERS
#Alter a few shops
> d[15:25,][1] = rpois(11,150)
> d[26,][4] = rpois(1, 10)
#look at the data
> head(d, 2)
       Romance Horror Science History
 A      36    107      62     139
 B      47     93      64     118
> books.PC.cov = prcomp(d)
> books.scores.cov = predict(books.PC.cov)
# Plot of PC1 vs PC2
> plot(books.scores.cov[,1],books.scores.cov[,2],
       xlab="PC 1",ylab="PC 2", pch=NA)
> text(books.scores.cov[,1],books.scores.cov[,2],labels=LETTERS)

This gives the following plot:

PCA plot http://img265.imageshack.us/img265/7263/tmplx.jpg

Notice that:

Shop z is an outlying point.
The others shops form two distinct groups.

Other possibilities

You could also look at GGobi, I've never used it, but it looks interesting.

Solved – How to visualize price distribution in marketplaces

For a dataset like that, I would start with histograms. First, do simple ones of price ranges, one histogram per keyword. Double check any outliers to make sure the data collection is correction (ie, you don't accidentally get an iPhone mixed in with your iPhone cases). Make sure the distributions make intuitive sense and there's no strange patterns for a given keyword.

Then, for each retailer, make a histogram of all of their products. Similarly, look for outliers or any issues you can spot. And then try to get sense of overall trends.

I might then create bar charts, one per retailer, with their range and average price across all products. How much do the ranges vary retailer to retailer ? It may be interesting to dig further into these by taking subsets of products (ie, who's cheapest for phone accessories ?) along with subsets of the retailers.

I would also want to add in some extra variables. For example, one that could be interesting is time. If you re-run the data collection each day for a couple of weeks, it would be interesting to look into the price fluctuations. Which retailers vary their prices the most ? What day of the week do you get the cheapest prices for each kind of product ?

That's a few initial ideas. As with any dataset, you will have to look at it a number of different ways and ask a number of different questions of it to find what's interesting and useful.

Best Answer

Related Solutions

Solved – Visualizing multiple “histograms” (bar-charts)

Solved – How to visualize price distribution in marketplaces

Related Question