Info

Swift Skunk
Author
Open Website

Testing and Issues

You can test this app and submit issues during the testing period of the Data Clustering Contest contest.

Entries with serious issues will not be able to win the contest, but even minor issues might be important for overall results.

Voting

148
by rating

Issues

Fair Leopard Feb 28, 2020 at 15:11
Final score for this submission (out of 100):

Languages: 95.24
News EN: 68.4
News RU: 65.31
Categories EN: 58.29
Categories RU: 67.67
Threads EN: 56.07
Threads RU: 49.58
Top news EN: 33.12
Top news RU: 37.13

These data reflect the relative accuracy, precision and speed of the algorithm as compared to the other submissions.
30
Fair Leopard Feb 6, 2020 at 16:03
In our preliminary tests, this submission received the following scores (out of 100):

Languages: 100
News EN: 85
News RU: 93
Categories EN: 78
Categories RU: 81
Threads EN: 78
Threads RU: 58
Top EN: 67
Top RU: 67

This is not the final result, please stay tuned for updates. We apologize for the delay.
20
Fair Mammoth Feb 7, 2020 at 20:38
В ходе предварительного тестирования алгоритма были выявлены следующие недостатки в ранжировании:

– Отсутствуют некоторые главные сюжеты в разделе ‘Main’ и внутри категорий. Заголовки части сюжетов слишком размытые (информация не подаётся в краткой нейтральной форме).

– Нарушена сортировка некоторых статей в сюжетах: релевантные статьи смешаны с нерелевантными.
20
Fair Leopard Dec 12, 2019 at 15:22
We had to fix the following issues before running the algorithm and will apply relevant penalties during the final scoring:
- invalid top category name, fixed "all" => "any"
10
Suave Duck Dec 12, 2019 at 21:40
Not pure articles similarity, except this - one of my favorites
2
Giant Croc Dec 12, 2019 at 22:18
Low precision for 'Other' in news detection, too many actual news weren't detected as news.
1
Swift Skunk Dec 13, 2019 at 03:27
Right, it can be improved for English. Although these articles are borderline news:
Black Friday rush:
https://data-static.usercontent.dev/sampledata/20191129/10/7193318762339737615.html
There is no actual event, just a photo of shoppers lining up with some comments.

Pokemon:
https://data-static.usercontent.dev/sampledata/20191129/10/5246707018890258376.html
The article is written almost as advertisement, which my filtering doesn't like.

If you scroll the "Other" part in News further, you will see that there are in fact very few articles that could be called genuine news.
Getahun Mesele Jan 17, 2020 at 12:56
In this article appears unknown English terms. https://data-static.usercontent.dev/sampledata/20191209/14/140598092120968031.html
Apple/safari/12.4
Nobody added any issues yet...