《信号与噪声》中涉及如下有趣又值得研究的问题:天气预报说降水概率为60%,你出门会带伞吗?被雷电击中的概率到底有多大?地震发生之前,我们真的无法预测吗?中情局为什么会忽略“9.11”恐怖袭击发生的信号?禽流感为何会突然爆发,又突然消失?为什么大数据时代的预测更容易失败?“信号”是我们想要和需要的事实,比如能帮助我们侦破早期鞋子炸弹案的信号。“噪声”则是另一回事,通常是不相干的信息,它阻碍或误导我们搜索信号。 “Signal and noise” involves the following interesting and worthwhile questions: The weather forecast says that the probability of precipitation is 60%, that you go out with an umbrella, and that the probability of being hit by a lightning bolt is great? We really can not predict it before the earthquake. Why does the CIA ignore the “9.11” signal of terrorist attacks? Why does the bird flu suddenly burst out and disappear suddenly? Why the prediction in the Big Data era is more likely to fail? “Signals” are what we want and need Facts, such as those that help us detect the early shoe bombing case. Noise is another, often irrelevant, information that hinder or mislead us in searching for signals.