Results · bigTwitter.json · final Spartan run
What 9 million tweets said about where Australia tweets from
The assignment asked three questions of a 18.7 GB file of geotagged tweets. These are the answers our MPI program wrote to data/result/ on Spartan in April 2023, transcribed from the submission (the Task 1 author IDs only as precisely as the report printed them). The raw tweets were course data and are not reproduced here.
- Tweets
- 9.09M
- 9,092,274
- Authors
- 119.4K
- 119,439
- Window
- 18 months
- Jul 2021 → Dec 20222021-07-05 → 2022-12-31
- In a capital city
- 74.7%
- 6,789,772 tweets
Task 2 · tweets per Greater Capital City
Melbourne edges out Sydney
Each tweet's place name was normalised and matched against the suburb gazetteer (sal.json) to find its Greater Capital City Statistical Area. Melbourne logged 66,220 more tweets than Sydney, and together the two account for 66.3% of all capital-city tweets.
Tweets per Greater Capital City
6,789,772 tweets (74.7% of all) resolved to a capital city. Rural areas such as 1rnsw are excluded, as in the original.
| Greater Capital City | Share of matched tweets | Number of tweets |
|---|---|---|
| Melbourne2gmel | 33.7% | 2,284,909 |
| Sydney1gsyd | 32.7% | 2,218,689 |
| Brisbane3gbri | 12.9% | 878,614 |
| Perth5gper | 8.7% | 590,045 |
| Adelaide4gade | 6.8% | 465,081 |
| Canberra8acte | 3.2% | 214,347 |
| Hobart6ghob | 1.3% | 91,112 |
| Darwin7gdar | 0.7% | 46,772 |
| Other Territories9oter | 0% | 203 |
Task 1 · most prolific authors
One account out-tweeted the next by 2.4 times
Tweets were counted per author_id on every rank, summed on rank 0 and ranked with ties sharing the best place. The top account posted 68,477 tweets in 18 months, about 126 a day.
Want to see the same reduction happen? Run it in the MPI lab.
Top 10 tweeters
task1.csv, all of bigTwitter.json
| Rank | Author id | Relative volume | Tweets |
|---|---|---|---|
| #1 | 1498063511204760000 | 68,477 | |
| #2 | 1089023364973210000 | 28,128 | |
| #3 | 826332877457481000 | 27,718 | |
| #4 | 1250331934242120000 | 25,350 | |
| #5 | 1423662808311280000 | 21,034 | |
| #6 | 1183144981252280000 | 20,765 | |
| #7 | 1270672820792500000 | 20,503 | |
| #8 | 820431428835885000 | 20,063 | |
| #9 | 778785859030003000 | 19,403 | |
| #10 | 1104295492433760000 | 18,781 |
Task 1 author IDs are shown exactly as published. The report table was pasted via a spreadsheet, which keeps only 15 significant digits, so the trailing digits of these 18–19 digit IDs read as zeros. The counts are unaffected.
Task 3 · authors across the most capital cities
10 authors tweeted from all eight capital cities
Authors were ranked by how many distinct capital cities they tweeted from, with ties broken by total tweets. Every top-10 author reached all eight, but in very different ways: the leader sent 98% of their tweets from Melbourne, while #5 spread theirs almost evenly across the country.
Tweets per city for the top 10 city-hoppers
Parsed from task3.csv; shading is each city's share of the author's tweets (hover a cell for the figure)
Swipe the table sideways to see all eight cities.
| # | Author id | SYD | MEL | BRI | ADE | PER | HOB | DAR | CTE | Total |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 1429984556451389440 | 11 tweets from Sydney, 0.6% of this author's | 1,879 tweets from Melbourne, 97.9% of this author's | 6 tweets from Brisbane, 0.3% of this author's | 2 tweets from Adelaide, 0.1% of this author's | 7 tweets from Perth, 0.4% of this author's | 1 tweets from Hobart, 0.1% of this author's | 1 tweets from Darwin, 0.1% of this author's | 13 tweets from Canberra, 0.7% of this author's | 1,920 |
| 2 | 702290904460169216 | 336 tweets from Sydney, 27.3% of this author's | 255 tweets from Melbourne, 20.7% of this author's | 235 tweets from Brisbane, 19.1% of this author's | 127 tweets from Adelaide, 10.3% of this author's | 156 tweets from Perth, 12.7% of this author's | 45 tweets from Hobart, 3.7% of this author's | 21 tweets from Darwin, 1.7% of this author's | 56 tweets from Canberra, 4.5% of this author's | 1,231 |
| 3 | 17285408 | 1,061 tweets from Sydney, 87.8% of this author's | 60 tweets from Melbourne, 5% of this author's | 40 tweets from Brisbane, 3.3% of this author's | 3 tweets from Adelaide, 0.2% of this author's | 7 tweets from Perth, 0.6% of this author's | 11 tweets from Hobart, 0.9% of this author's | 4 tweets from Darwin, 0.3% of this author's | 23 tweets from Canberra, 1.9% of this author's | 1,209 |
| 4 | 87188071 | 116 tweets from Sydney, 28.5% of this author's | 86 tweets from Melbourne, 21.1% of this author's | 68 tweets from Brisbane, 16.7% of this author's | 28 tweets from Adelaide, 6.9% of this author's | 52 tweets from Perth, 12.8% of this author's | 15 tweets from Hobart, 3.7% of this author's | 5 tweets from Darwin, 1.2% of this author's | 37 tweets from Canberra, 9.1% of this author's | 407 |
| 5 | 774694926135222272 | 37 tweets from Sydney, 13.6% of this author's | 38 tweets from Melbourne, 14% of this author's | 37 tweets from Brisbane, 13.6% of this author's | 28 tweets from Adelaide, 10.3% of this author's | 34 tweets from Perth, 12.5% of this author's | 36 tweets from Hobart, 13.2% of this author's | 28 tweets from Darwin, 10.3% of this author's | 34 tweets from Canberra, 12.5% of this author's | 272 |
| 6 | 1361519083 | 18 tweets from Sydney, 6.8% of this author's | 36 tweets from Melbourne, 13.5% of this author's | 1 tweets from Brisbane, 0.4% of this author's | 9 tweets from Adelaide, 3.4% of this author's | 1 tweets from Perth, 0.4% of this author's | 2 tweets from Hobart, 0.8% of this author's | 193 tweets from Darwin, 72.6% of this author's | 6 tweets from Canberra, 2.3% of this author's | 266 |
| 7 | 502381727 | 2 tweets from Sydney, 0.8% of this author's | 214 tweets from Melbourne, 85.6% of this author's | 8 tweets from Brisbane, 3.2% of this author's | 4 tweets from Adelaide, 1.6% of this author's | 3 tweets from Perth, 1.2% of this author's | 8 tweets from Hobart, 3.2% of this author's | 1 tweets from Darwin, 0.4% of this author's | 10 tweets from Canberra, 4% of this author's | 250 |
| 8 | 921197448885886977 | 49 tweets from Sydney, 23.7% of this author's | 56 tweets from Melbourne, 27.1% of this author's | 37 tweets from Brisbane, 17.9% of this author's | 24 tweets from Adelaide, 11.6% of this author's | 28 tweets from Perth, 13.5% of this author's | 4 tweets from Hobart, 1.9% of this author's | 1 tweets from Darwin, 0.5% of this author's | 8 tweets from Canberra, 3.9% of this author's | 207 |
| 9 | 601712763 | 44 tweets from Sydney, 30.1% of this author's | 39 tweets from Melbourne, 26.7% of this author's | 11 tweets from Brisbane, 7.5% of this author's | 19 tweets from Adelaide, 13% of this author's | 14 tweets from Perth, 9.6% of this author's | 8 tweets from Hobart, 5.5% of this author's | 1 tweets from Darwin, 0.7% of this author's | 10 tweets from Canberra, 6.8% of this author's | 146 |
| 10 | 2647302752 | 13 tweets from Sydney, 16.3% of this author's | 16 tweets from Melbourne, 20% of this author's | 32 tweets from Brisbane, 40% of this author's | 3 tweets from Adelaide, 3.8% of this author's | 4 tweets from Perth, 5% of this author's | 5 tweets from Hobart, 6.3% of this author's | 3 tweets from Darwin, 3.8% of this author's | 4 tweets from Canberra, 5% of this author's | 80 |
Raw output
The result files, as published
Rank 0 wrote Task 1, rank 1 wrote Task 2 and rank 2 wrote Task 3, each as a CSV. They are laid out here in main.py's format, header typo (“Captical”) included, because that is also the format the browser port writes. Task 2 and Task 3 are as submitted. The Task 1 author IDs went through a spreadsheet on the way into the report and lost their trailing digits; main.py wrote them in full, and the counts are exact.
Reconstructed from the report. Task 1 author IDs are shown exactly as published. The report table was pasted via a spreadsheet, which keeps only 15 significant digits, so the trailing digits of these 18–19 digit IDs read as zeros. The counts are unaffected.
Rank,Author Id,Number of Tweets Made
#1,1498063511204760000,68477
#2,1089023364973210000,28128
#3,826332877457481000,27718
#4,1250331934242120000,25350
#5,1423662808311280000,21034
#6,1183144981252280000,20765
#7,1270672820792500000,20503
#8,820431428835885000,20063
#9,778785859030003000,19403
#10,1104295492433760000,18781