纪录片《为何无法预测地震》 第22期:预测的准确性（在线收听）

Sometimes, humans can see the patterns and we propose

有时 我们可以观察到那些模式

something that seems to us logical

并且按照地震发生的方式

in terms of the way earthquakes behave,

按照我们的思维逻辑来提出建议

but sometimes, their patterns are too complicated

然而有时 地震的模式太过复杂

and the hope is that computers,

人们就只能希望借助计算机

using vast amounts of data,

利用大量的数据

and, er, combing the data for those patterns,

然后将数据和那些模式结合

can out-think us in that particular way.

代替我们计算出结果

But the patterns haven't always

然而 研究地震的模式

led to accurate predictions.

并不总能作出准确的预测

Nine years after Northridge,

在北岭地震发生后9年

Keilis-Borok's team announced

凯列斯勃洛克小组宣布

that a major earthquake would strike near Palm Springs

在2004年9月5日 美国棕榈泉附近

by September 5th, 2004.

将发生一次大地震

Once again,

再一次

the enigmatic Russian was putting his career on the line.

这神秘的俄国人赌上了他的整个职业生涯

But this time, nothing happened.

然而这一次 地震没有发生

The team's work continues to be

小组的研究工作

a mixture of success and failure,

继续徘徊于成功和失败之间

but Keilis-Borok is confident

但是凯列斯勃洛克坚信

that he can improve his hit rate.

他的预测准确率可以提高

There is no such thing

做到百分之百准确

as 100% accuracy,

根本不可能

It's go without exist, but

那是无稽之谈 但是

we believe the accuracy can be increased

我们认为预测的准确性

by a factor of at least five.

至少可以提高五个点