I came to understand this from my experience as chief executive of PayPal. In mid-2000 we had survived the dotcom crash and we were growing fast but we faced one huge problem: we were losing upwards of $10m a month to credit card fraud. Since we were processing hundreds or even thousands of transactions each minute, we could not possibly review each one. No human quality control team could work that fast.
我是从担任Paypal首席执行官的经验中得出这一结论的。在21世纪头十年中期,我们挺过了互联网泡沫破灭,得到了快速增长,但我们遇到了一个大问题:我们每月因信用卡欺诈起码要损失1000万美元。由于我们每分钟要处理成百上千笔交易,所以不可能对每笔交易进行审查。由人力构成的质量管理团队绝不可能审查得这么快。
We tried to solve the problem by writing software that would automatically identify bogus transactions and cancel them in real time. But it quickly became clear that this approach would not work: after an hour or two, the thieves would catch on and change their tactics to fool our algorithms.
我们试过用编写软件来解决这一问题,该软件可以自动识别虚假交易,并实时取消。但事实很快证明这种方法行不通,只一两个小时后,窃贼们就搞懂了软件,然后改变策略以欺骗我们的算法。
Human analysts, however, were not easily fooled by criminals’ adaptive strategies. So we rewrote the software to take a hybrid approach: the computer would flag the most suspicious transactions, and human operators would make the final judgment.
【造福社会影响文明 你不必与科技为敌】相关文章:
★ 永远不要以貌取人
最新
2020-09-15
2020-09-15
2020-09-15
2020-09-15
2020-09-15
2020-09-15