Auto Byte

专注未来出行及智能汽车科技

微信扫一扫获取更多资讯

Science AI

关注人工智能与其他前沿技术、基础学科的交叉研究与融合发展

微信扫一扫获取更多资讯

Yazmin How, Digital Content Manager - RE•WORK 作者

AI是如何粉碎金融诈骗的呢?

机器学习和人工智能正在彻底改变世界。 他们能够降低成本,分析数据,识别人眼无法看到的模式和趋势,并做出实时决策。 现在,他们被用来帮助防止金融欺诈,他们每天都在学习如何做得更好。 目前估计网络犯罪使全球经济损失约6000亿美元,其中最常见的形式之一是信用卡欺诈,随着在线市场的增加,信用卡欺诈已大幅增长。 随着越来越多的人选择在线交易等金融服务,更好,更快,更准确的欺诈检测和预防技术变得越来越重要。

Original

Machine learning and artificial intelligence are revolutionising the online world. They are capable of reducing costs, analysing data, recognising patterns and trends we can’t see with the human eye and making real-time decisions. Now, they are being used to help prevent financial fraud and they’re learning how to do it better every day.

Currently it is estimated that cybercrime costs the global economy approximately $600 billion, with one of the most common forms being credit card fraud which has grown considerably with the increase in the online market. As more and more people chose to transact online it is becoming increasingly important for financial services to invest in better, faster and more accurate fraud detection and prevention techniques.

How our data helps protect us

Thanks to there being such a large amount of online transactions, this means that there is a huge amount of customer data available which can be studied and learnt by AI. They can learn how to identify valid credit card behavioural patterns and how to detect irregular behaviour which could be fraudulent.

Identity theft is just one of the types of fraud which has plagued the financial services. Many services now use a translytical database which performs real-time fraud analysis when you use a credit card and make payments. This means that every time you make a payment the system learns your behaviour, if this behaviour changes or shows something unusual, the system will flag it as being fraudulent.

“These models are constantly learning and being trained with real data from fraudulent transactions and real transactions so that they can better protect you. Of course, false positives are still possible but the more the system learns, the less frequent these results become,” says Mavis Hucks, chief analytics officer and writer for Last Minute Writing and Researchpapersuk.

Multi-layered defences led by AI

In order to achieve the best protection for their customers, financial services often adopt a multi-layered technique, using multiple different tools including AI to deter fraudsters and protect customers. Cybersecurity companies are blossoming as their services are vital to keep businesses, customers and financial services safe. Their algorithms have improved over the last few years to help increase their accuracy.

AI is learning human behaviour at a level which understands when and where you use your credit card. For example, if you used your card abroad it would often get flagged as fraud. Now the AI has learnt how to detect an underlying pattern, it can see that people will often buy a plane ticket, accommodation and then make a transaction in a foreign country. It will compare this data to previous transactions in order to decide if the transaction should be successful or be declined.

Reducing incorrect results through learning

This not only helps to protect us, but it also increases customer satisfaction by reducing the number of false positives produced by the algorithm. After all, no one likes to have their card declined when all they want is to relax on vacation!

The evolution of anti-fraud

The future of fraud detection is looking even brighter thanks to a new generation of algorithms known as Convolutional Neural Networks which are based on the way people think. They are designed on the visual cortex and use images directly as input. “The neural networks study the relationships between a user and their spending at a much deeper level than ever before. Once this technology is widely implemented cybercrime is likely to drastically be reduced,” adds Blain Meserve, senior principal consultant at Draftbeyond and Writinity.

AI provides a win-win situation for the financial sector and for anyone who uses a credit card. It saves both the banks and their customers huge costs every year. As it learns and understands more about us and the criminals, it becomes better at protecting us. It is hoped that the AI technology will continue to become more advanced and will crush financial fraud in its wake.

Martina Sanchez is an entrepreneur and content marketing specialist at Lucky Assignments and Gum Essays. She is absorbed with article writing and is a constant contributor to her blog where she touches such topics as digital marketing, SEO tips and tricks etc.

Original

RE•WORK
RE•WORK

RE•WORK成立于2013年,宗旨为促进国际人工智慧及相关之研究,发展,应用及交流。迄今为止,RE•WORK在世界各地已创办超过50场人工智能高峰会,包括新加坡,香港,美国,伦敦,加拿大等等。

产业AIFinanceDeep Learning in Finance SummitFintech
1
暂无评论
暂无评论~