Hacker News Reader

malkungon Ask HN: Why use Python or similar?

It is much easier on your eyes. Having to scroll through code bloat is just physically painful.

malkungon Tokenising the english text of 30TB common crawl

I once tried to identify the language in (supposedly) English-language Facebook posts. The second-most frequent language was Estonian. Looking at actual posts classified as Estonian, it turned out they contained text like "Soooooo nice". Because the language identifier looks at character sequences, and Estonian tends to have unusually common double vowels, such posts were classified as Estonian. Dutch and Norwegian were other two languages for which English was mistaken - apparently some character sequencies there have frequency distributions that are similar to English.

malkungon Russian legislative elections 2011 - statistical evidence of vote fraud

What is "weighted"? In which way was the percentage weighted? Why look at (weighted) percentage of turnout rather than at raw counts?

malkungon Russian legislative elections 2011 - statistical evidence of vote fraud

Because bureaucracy has grown a lot under Putin, and those people might as well lose jobs if some other party comes to power. Also think about increases in salaries and pensions, before the elections, etc. But these do not count as direct vote-rigging.

malkungon Russian legislative elections 2011 - statistical evidence of vote fraud

Another explanation is that people working in government bodies in Russia (including state-funded education, healthcare and military institutions) make up some 30 million, so United Russia gets a huge number of votes just from them. Moscow has a large army of bureaucrats, you know.

malkungon Russian legislative elections 2011 - statistical evidence of vote fraud

Chechnya usually votes close to 100% for Putin, simply because a lot of government money is poured in Chechnya, disproportionately more than in any other region of Russia,

malkungon Russian legislative elections 2011 - statistical evidence of vote fraud

Interesting. Did you look at the polling stations at the irregular spikes? What is their number, rather than percentage. The absoulte number (even better, the number of voters in them) would indicate what effect hey had on the overall results. Also what regions are they in? In some very sparsely populated regions, like North Siberia, where "mobile" polling stations are used, the number of voters would be very small, but probably the attendance would be close to 100%, and I would not be surprised if they all voted for United Russia. Or, this may be due to some other peculiarity of counting what constitutes a polling station (stations abroad, the voting mechanism for people who were travelling on the day of the elections, etc). Also, please do confirm where the data comes from - is it publicly available?