What is machine learning? Artificial intelligence? Machine learning? Deep learning? It is difficult to keep track of what all these terms mean and how they connect to one another. While they each have specific definitions and represent different specialty areas, there is overlap and all of them are linked to the overarching pursuit of enhancing computer abilities.
Artificial intelligence is a broad term that covers the overall study of advancing computer capabilities to the point of matching human intelligence including the ability to reason, plan, learn, and physically move
Machine learning is typically viewed as a subset of artificial intelligence as it represents a computer’s ability to learn (obviously, a key component in creating true artificial intelligence). It is the study of having computers learn by leveraging data inputs and algorithms to find correlations, without relying on specific programming based on rules.
Supervised learning involves a “teacher” that guides the computer by loading it with inputs and desirable outputs with the goal of having the computer learn how to match inputs with those outputs
Unsupervised learning involves inputs are that uncharacterized and forces the computer to learn which characteristics of the inputs are linked to outputs
Deep learning is a branch of machine learning and involves the construction of algorithms that take data input and transform them to deliver an output, which is then used as a subsequent input to a new algorithm resulting in several layers of inputs and outputs. It has also been referred to as neural networks.
Natural language processing is a type of deep learning that uses text as an input and attempts to infer the meaning of a combination of text, such as a question or command
There are several companies that are either experimenting with machine learning or integrating it into their core offering. Of course, big technology companies like Google have been active in this field for some time. In 2015, it was announced that Google was adding RankBrain into their search system to learn how to best respond to incoming queries. Though there are clear applications of machine learning in the tech space, there are many companies involved in the financial services industry that have found interesting use cases for this technology.
Interesting AI/ML firms in financial services Many companies involved in the financial services industry are actively exploring the machine learning technology. Some interesting companies that are active in this space include:
Ayasdi is a machine intelligence and big data software company that offers companies their platform to analyze large amounts of data and build predictive models. Their most common use cases are in the healthcare and financial services areas. Specifically, they offer solutions for anti-money laundering, fraud detection, and market regime forecasting. Kabbage leverages machine learning to power their lending platform, which is able to analyze credit risks and consistently monitor transactions. As more consumers and small businesses leverage their platform to receive loans, more data is used to understand trends and patterns. This, in turn, enhances their risk scoring and predictive capabilities. Dataminr is a big data company that uses machine learning in their analysis of social media and publicly available datasets. With their ability to digest large amounts of data from sources such as Twitter, they are able to understand trends as soon as they begin emerging. This information is passed on to their customers, many of which are in the financial markets, in real time so they can make more informed decisions, such as how to update their trading strategy. Pefin is a financial advising company based on artificial intelligence technology that will be launching later this year. Using neural networks, they will deliver financial planning advice to their customers based on millions of data points, many of which are specific to the unique end user. WealRo is a startup that seeks to be the “Siri for saving and investing” by integrating machine learning into their offering. They are in the final stages of launching their product which will be a voice-activated savings and investment solution that will give suggestions for budgeting and will offer the ability to trade stocks.
Key opportunities for financial services firms As machine learning capabilities become more and more robust, there will be more opportunities for banks, hedge funds, and insurance companies to transform the way they do business. Everyone has been talking about big data, but machine learning is the true path to actually changing processes and decision-making into a smarter and more advanced state. We have identified four key opportunities for financial services companies:
Risk underwriting processes: banks distributing loans and insurance companies offering coverage policies all revolve around predicting and controlling risks. Up until this point, they have relied on human experts to analyze historical data and develop models to price and score these risks. By leveraging machine learning, more advanced and accurate models can be created to help protect these companies from risky investments.
Customer targeting and segmentation: from a sales and marketing standpoint, machine learning can greatly improve overall performance. Over the past couple of decades, marketing has shifted from a creative profession to one more focused on data and analysis. Machine learning is a natural progression in this evolution, as these models will help to predict which market segments are underserved and what methods work best with specific consumer groups.
Fraud detection: similar to developing advanced actuarial and risk scoring models, it will soon become more accurate and efficient to utilize machine intelligence in the development of fraud detection methods as opposed to using rules built by human experts. As we continue to collect increasing amounts of data on consumer behavior, it will be easier for computers to flag transactions that fall outside of the recognized patterns.
Investment and valuation strategy: wealth management firms, hedge funds, and investment banks will all benefit greatly by the growth in machine learning advancements. Current robo-advisors simply leverage advanced algorithms based on human-defined rules. But in the future, machine learning algorithms will help detect subtle market trends and will be used to optimize portfolios and value private companies based on a multitude of factors.
While we may be far off from having true artificial intelligence like we’ve seen depicted in countless movies, there has been dramatic progression in technological capacity and that trend will surely continue. Machine learning is a clear example of that progression and there is a lot of potential to use this technology in our industry. Whether you focus on risk, marketing, or trading there’s a reason to be excited and a reason to start focusing on this space!
What are your views on machine learning in financial services? Email us, we would be happy to have your insight.
Happy Reading! The CH&Co. Editorial team
The Latest AI / ML Articles
Clever banking with artificial intelligence
The co-founder of Market Gravity, Gideon Hyde, describes why and how he believes artificial intelligence and machine learning will be integral in staying competitive in banking and financial services. Read the full article here.
This Is The Future Of Artificial Intelligence
The CEO of Xerion Investments, Daniel Arbess, discusses his view on the trends that artificial intelligence is undergoing and how that will influence the direction of our global economy. Read the full article here.
Wall Street hopes software helps it hire loyal bankers
Some firms on Wall Street are using software with smart algorithms and artificial intelligence to make hiring decisions to ensure they attract the right kind of talent. Read the full article here.
Artificial intelligence, risks and regulation in financial services Several industry experts share their insights and perspectives on how artificial intelligence will influence the global economy and financial services in the coming future. Read the full article here.
CH&Co Digital Case of the Month: ThetaRay
CH&Co. co-founded FinFusion, the first non-for-profit Canadian organization that helps develop, promote and support innovation in the finance and technology community (FinTech) in Montreal, in Quebec and in Canada last June. The launch cocktail reuniting Montreal’s FinTech community will take place the 5th of July.
CH&Co. attended the Fin&Tech Community Conference on 9th June 2016 in Paris. We participated as a sponsor of the event and met several different startups in insurance, crowdfunding and digital health as part of the Speed Networking events throughout the day.
CH&Co. is co-organizing the NextMoney “InsureTech Future – Future customer behavior and the IOT Opportunity” meetup in Singapore on July 28, 2016. The event will bring leaders from insurance companies, fintechs and other innovators featuring panel discussion, fintech demos and key insights from CH&Co.’s connectivity IoT benchmarking.
Conferences, exhibitions, workshops…If you would like to participate in our events, please register here. Membership is free!
Share the news!
If you enjoy CH&Co.’s news, please share it with your colleagues. They can also sign up for future editions. Membership is free!
Chappuis Halder & Co. is a consulting firm specialized in Financial Services with offices in North America, Europe and Asia. We help our clients in several industries, Corporate & Investment Banking, Commodity Trading, Insurance and Retail & Private Banking, with a permanent focus on expertise and research, especially in the Digital area.
Privacy & Cookies Policy
Necessary cookies are absolutely essential for the website to function properly. This category only includes cookies that ensures basic functionalities and security features of the website. These cookies do not store any personal information.
These cookies allow us to determine the number of visits and sources of traffic to our website in order to measure and improve performance. They also help us to identify which pages are most / least visited and to evaluate how visitors navigate through the site. All information, collected by these cookies, is aggregated and therefore anonymized. If you do not accept these cookies, we will not be able to know when you have visited our website and we will not be able to measure its performance.