Machine learning, deep learning, and AI are all buzzwords for good reason: these technologies are fundamentally shifting the nature of business, society, and our lives. More importantly, across many verticals, they’re shifting from being disruptive technologies to being foundational and table stakes for businesses to remain competitive.
THE POWER OF MACHINE LEARNING ACROSS VERTICALS
To get started, let’s look at several key examples of how machine learning (ML) is impacting various verticals:
- In tech, ML powers recommendation systems, content discovery, search engines, email spam filters, and matching problems.
- In healthcare, machine learning facilitates drug discovery and diagnostic imaging diagnosis.
- In finance, ML is now foundational for fraud detection, process automation, algorithmic trading, and robo-advisory.
- In retail, Walmart is at the forefront of using ML to reinvent supply chain management.
- Burgeoning industries such as LegalTech and AgTech (Agriculture Technology) are growing rapidly and employing machine learning. Legal technologists are imagining a future in which ML is leveraged to predict outcomes of court cases based on natural language analysis of precedents. And in agriculture, drones are being deployed at scale to capture footage, and ML is being used to estimate crop yields.
THE POWER OF MACHINE LEARNING ACROSS TEAMS
The above examples are vertical-specific, but there are also many gains in efficiency in the development of ML algorithms that are vertical-independent, such as:
- HR teams filtering applicants in the hiring flow.
- Support teams using ML for call center routing.
- Marketing teams using ML algorithms for paid advertising, customer churn prediction, and targeted nurture campaigns.
In fact, any company that has an app can benefit from leveraging ML to determine the most effective push notifications, and any organization that has a website can leverage ML to personalize their customer experience by surfacing content and features that are most relevant to them.
According to Gartner’s Annual Chief Data Officer Survey, poor data literacy is the second-biggest internal roadblock to the success of chief data officers. Gartner expects that, by 2020, 50% of organizations will lack sufficient AI and data literacy skills to achieve business value, and 80% of organizations will initiate targeted data literacy initiatives to overcome deficiencies. The data is clear: To keep your competitive advantage, you’ll need to leverage ML in one form or another. The question is, as a business leader, what do you need to know about it?
HOW TO SLICE DATA ANALYTICS
The first thing to do is to place ML in the broader context of what data analytics and data science can offer. One instructive way of slicing the data science space is into the following areas:
Descriptive analytics is essentially about getting the right pre-existing data in front of the right people in the form of dashboards, reports, or emails. This can include both past and real-time time data about revenue, customer engagement, churn, and company and employee performance.
Predictive analytics is synonymous with ML and is the realm of predicting the future (such as whether a customer will churn or not) and more general classification tasks (such as whether an email is spam or not, and if a tumor is benign or malignant).
Prescriptive analytics is the realm of decision science and how to make decisions based on data. If your ML model tells you that a particular customer will churn, you’ll want to know what to do about it. Prescriptive analytics is concerned with finding frictionless interfaces between the data and decision functions in any organization. Exciting spaces to watch are data translation (a burgeoning field for those with both domain expertise and technical know-how), advances in reinforcement learning (which bleeds into ML; see below), and the work of Cassie Kozyrkov, Chief Decision Scientist at Google Cloud, with whom I discussed decision science on DataFramed, the DataCamp podcast.
TYPES OF MACHINE LEARNING
Now that we have a sense of the data space, let’s dive into the different types of machine learning:
- Unsupervised learning is about discovering general patterns in data, the most popular example being clustering or segmenting customers and users. This type of segmentation is generalizable and can be applied broadly, e.g. to documents, companies, and genes.
- Supervised learning currently enjoys the lion’s share of ML, and is concerned with the prediction and classification of data.You may notice that when people talk about ML, they’re probably specifically focused on supervised learning.
- Reinforcement learning is concerned with training ML models to make decisions. Examples include many promising algorithms for self-driving cars and AlphaGo Zero, which was not trained on any human game data and was able to achieve superhuman performance in Chess, Go, and Shogi after 24 hours of training.
In the short to medium term, supervised learning will be the most important ML technique for business leaders so we’ll focus on it in subsequent posts. However, do keep reinforcement learning on your radar.
← Back to blog