Niklas Fejes
100/100
6 hours
Sunday
machine learning, challenge 2
- Using
d3.svg.line
to create plots. - More advanced usages of
d3.svg.axis()
, such as including units and custom labels
> db.aiddata.find({donor:'Belgium'},{recipient:1, disbursement_amount:1})
> db.aiddata.find({biodiversity:{$gt:0}},{recipient:1,disbursement_amount:1,title:1})
> db.runCommand({distinct: 'aiddata', key: 'flow_type', query: {}})
> db.runCommand({distinct: 'aiddata', key: 'flow_type', query: {disbursement_amount:{$gt:100000000}}})
In the query, stats.n
is the number of elements used to generate the distinct values. When working with all
values this number was 1000859 (all values in the dataset) and with the disbursement limit it was 994, which is
the number of entries with disbursement_amount
larger than 100'000'000.
> db.aiddata.aggregate([ {$match: {donor:'Belgium'}}, {$group: {_id:"$year", total:{$sum: "$disbursement_amount"}}} ])
It may be interesting to know whether the precence of the bugs affects the pitch of the phone. Plotting the pitch for each user vs time gives the following plots.
In the plots above, the flat parts are the time when surveys were entered. To get a proper analysis of the different game stages, one would have to separate these into different "events".
What can be seen is that for some of the users, the third game appears to have more "spikey" data. This could be an indication of the bugs in the game.
var query = {"city":"Middleton"};
var query = {"city":"Las Vegas"};
This graph has the following properties:
selection: Businesses in Las Vegas
x-axis: number of ratings
y-axis: latitude
color: number of stars
It shows that the most rated businesses are in the middle of the city, and that the average business rating seem to be even across the town.