Your single source for new lessons on legal technology, e-discovery, and the people innovating behind the scenes.

4 Ways You Trust Machine Learning

Sam Bock

According to a recent survey, 93 percent of corporate legal departments believe analytics will be critical to their work in the next ten years. Evidence of this demand is clear: Thousands of gigabytes of data are pushed through Relativity Analytics each month, and that number is climbing at an incredible pace. Still, all of that amounts to fewer than 11 percent of cases using analytics features. Why is that?

In many cases, it’s about trust. Attorneys are trained  to limit risks—their clients’ livelihoods or organizations’ reputations are often at stake, after all. The judicial approval of technology-assisted review in courtrooms around the globe can be quite helpful in addressing this skepticism, as can insights from technology-minded practitioners who are passionate about making e-discovery more efficient.

But you know what else might help? Your credit card statement.

We trust machine learning to make our lives easier every day—for something as unimportant as which movie to watch next, to critical necessities like preventing credit card fraud. How has this technology already made your life more productive?

1. Money Management

Machine learning and predictive analytics can mean the success or failure of your 401(k). The projected growth of your investments comes not just from the current rates, but analytics performed on the funds’ historical rates and, sometimes, totally external factors—presidential elections are often responsible for changes in the stock market, for example.

Additionally, the fraud detection and prevention programs that keep your bank account safe are also utilizing predictive analytics. The most sophisticated of these programs analyze your spending habits and compare each purchase against them. The purchases themselves also have fraudulence probability scores (a $1,000 online purchase paid to a company based in Timbuktu is more likely to be fraudulent than a $200 purchase at your local grocery store). If anything seems fishy, the bank sends you an alert to ensure your funds aren’t compromised if fraud is, in fact, taking place.

Predictive analytics also help insurance companies set fair rates and evaluate risks. The result is a cost-effective insurer who can keep your family financially protected in an emergency.

2. Online Experiences

Just as the internet is home to big data, it’s also home to smart data. Companies like Google and Facebook literally run on ad revenue—and their ads are successful because they are strategically placed in front of the right users. Ever noticed how a Google search for crock pot recipes later yields pervasive ads for Bed Bath & Beyond and Whole Foods? (When this author was planning her wedding, the number of ads she saw for caterers, venues, and photographers was almost unbelievable.)

Similarly, your online shopping is probably made easier by listings for “items you might like” on websites like Amazon. That uncanny suggestion for a GoPro that’s compatible with that sweet new drone you bought a few weeks ago isn’t there by chance, after all.

Another example affecting your online life is the collection of junk, spam, and clutter filters on your inbox. Email providers are getting increasingly sophisticated at knowing what you want to see and what you don’t—and it’s not based solely on simple rules about sender aliases anymore. You’ll notice these filters will change over time based on the actions you take on incoming messages.

3. Streaming Entertainment

One go-to analogy our training team uses to explain Relativity Assisted Review to new users is to compare the workflow to popular online music service Pandora. Pandora works by enabling users to create their own radio stations. As each song begins, the listener is able to provide a “thumbs up” or a “thumbs down.” Over time, the stations are better trained to serve up music the listener will enjoy.

Similar concepts are responsible for the suggested movies and TV shows that show up in your Netflix account. The recommendations you see as the credits roll on the latest series finale you’ve completed aren’t coincidental, and they aren’t based solely on the objective similarities between programs. They’re also based on what you watch and enjoy. All the better for enabling your weekend binge-watching sessions.

Technology-assisted review offers a similar experience, except that instead of training the computer on your favorite tunes, you’re training it on the most relevant documents so the system can identify the similarly relevant data in the rest of the set. Simple, right?

4. Travel Expenses

Before planning a big trip, most of us do a lot of price comparison on flights and hotels. We all know it’s more expensive to travel on holidays like Thanksgiving than it is on a random Monday in January. But why are tickets often less expensive on Tuesdays? How do we know that 40-60 days in advance is the ideal time to buy?

You got it: Analytics. Airlines and hotels use their own data to set pricing—as well as flight schedules—based on the predicted demand they can glean from past purchase data. Fortunately, consumers can get in on the analytics game by analyzing pricing data to help them determine the best time to book.

Although the intricacies behind these processes differ from tools purpose-built for e-discovery, their goals are the same: Machine learning assists human efforts by helping you understand, access, and act on the best pieces of massive data sets more quickly. TAR works behind the scenes to learn what you’re looking for in a hot document and help you categorize records based on what’s most and least likely to fall into that realm of responsiveness.

So next time you or someone on your team doubts the efficacy of data analytics for your e-discovery projects, consider all the ways you trust strikingly similar technology to improve your day-to-day life. Although the risks and considerations at play aren’t the same, the exercise could spark some imaginative—and entertaining—discussion on how your experience with data in all areas of life can improve.

What other applications of machine learning improve your life? Let us know in the comments or @kCura on Twitter.

Sam Bock is a member of the marketing team at Relativity, and serves as editor of The Relativity Blog.

The latest insights, trends, and spotlights — directly to your inbox.

The Relativity Blog covers the latest in legal tech and compliance, professional development topics, and spotlights on the many bright minds in our space. Subscribe today to learn something new, stay ahead of emerging tech, and up-level your career.

Interested in being one of our authors? Learn more about how to contribute to The Relativity Blog.