How Machine Learning Can Bring Overnight Change to Businesses

Move over Darwin. Let’s consider how machine learning (ML) can demonstrate evolution-style growth — but with a difference in speed.

First, ML is adaptive, always responding to data input. It continues to grow on its own like some parallel world that has absorbed our human world and reflects it back to us. Because ML is trained to spot patterns in data, its self-training capacities “select” for patterns (without losing sight of anomalies), organizing our world around meaning. ML becomes so good at pattern discovery that it sees patterns within patterns and patterns among disparate fields.

What’s more, ML like Salesforce Einstein captures subtle variation by building individualized models for every customer’s unique set of data, workflows, and processes. Add ML’s speed to the mix, thanks to its massive data crunching, and it’s easy to see how deep-and-wide growth takes off.

This speed sets ML apart. Although it’s iterative, ML is not in the slow lane of incremental change seen in nature. Machine learning is overnight change, sweeping industry change.

In “The Great A.I. Awakening,” an article in the New York Times Magazine, Gideon Lewis-Kraus shows just how fast and powerful ML’s impact can be. He notes that Google Translate shifted from a statistical methods technology to an A.I.-based system. Within a day of this switch, Japanese people participating in a social media quiz couldn’t tell whether a passage from a Hemingway novel (“The Snows of Kilimanjaro”) was translated into English by a Japanese professor or by Google Translate. Thanks to ML, Google Translate immediately began to process and produce language with such elegance and naturalness that the ML powering it passed the famous “Turing Test” — a computer successfully deceived humans.

Lewis-Kraus goes on to explain how major companies, like Google and Microsoft, are advancing their ML tools and platform services because they are jockeying to define the radical change we’re in the midst of and don’t fully see.

“What is at stake is not just one more piecemeal innovation but control over what very well could represent an entirely new computational platform: pervasive, ambient artificial intelligence,” said Lewis-Kraus.

Put another way, users may have once been freaked out by Facebook’s early use of ML in its recommendation engine. That now looks like child’s play compared to the strange new behavioral world we’re entering with Meta. With virtual reality, we’ll soon be able to immerse ourselves in a new culture of trying out things — furniture, real estate, hotels, and travel experiences — before deciding to purchase. How will these heightened bespoke customer experiences impact the user? Stay tuned.

In its own way, Salesforce has revolutionized the customer relationship management (CRM) industry with its easy-to-use programs and its ML-driven applications that drive continuous improvement. To take just one example: automation and deep personalization go hand in hand, perhaps ironically, as we march into the future of rich customer experiences.

Here’s a look at some industries harnessing the power of ML and what it’s enabled them to do. Keep in mind that with ML, it’s never too late to jump in, no matter how big or small your business is. You can be a company who is plodding along without ML, and then build or integrate ML into your business processes, and you’ll quickly see its impact. Performance, profit, growth, and brand loyalty are a few of these striking benefits.

Customer Relationship Management

From retail and hospitality to the food and beverage industries, companies increasingly depend on CRM software that uses machine learning models to strengthen customer relationships and business. Whether sales, marketing, or operations teams are out in the field, in cyberspace, or the office, machine learning software like Salesforce Einstein can help businesses:

• automate decision-making, like the best action to take
• resolve HR issues through automation
• better track customers’ purchasing history
• understand customers at a deep level by extracting patterns, stats, and other customer behavior
• predict customer behavior
• generate leads
• find information fast to have answers & insights at a team member’s fingertips
• classify customer cases
• analyze and prioritize emails
• craft email responses
• generate revenue forecasting
• tailor product development and marketing initiatives based on customer data

Business Processes and Business Intelligence

Industries ranging from banking, finance, and insurance to smart energy, automotive, and telecommunications are using ML to:

• codify knowledge to automate decisions
• improve business processes and operations
• predict business expenses and perform cost analyses to increase profitability
• improve analytics, including real-time data analysis and spotting anomalies as well as patterns
• provide forecasting answers; ML uses massive historic data to produce accurate estimates of future behavior
• improve data quality checks; for instance, DataGroomr, Salesforce’s ML-based deduplication app, is trained to spot duplicates, cleansing data continuously without the need for ongoing human intervention
• engage in algorithmic trading, known as automated trading. By programming a set of instructions for the trade (based on price, quantity, etc.), the computer can execute the trade at lightning speed. Investors and stockbrokers use ML to predict market conditions, mitigate emotional bias, and increase profits
• risk management
• gain a greater understanding of the overall business and consumer needs

Fraud Detection

Consumers reported losing more than $5.8 billion to fraud in 2021 — an increase of more than 70 percent over the previous year, according to the Federal Trade Commission. The use of technology in the finance industry – particularly with digital payments and transactions – has led to more fraud, scams, and phishing. To fight back, companies are using ML to:

• analyze consumers’ current patterns and transaction methods for instantaneous detection of a behavioral difference; in real-time, users can confirm approval to complete a transaction.
• increase the accuracy of fraud detection, compared to human analysis
• save on costs since massive amounts of data are analyzed in milliseconds without the need for manual review
• improve detection of email phishing with precise analysis and classification of good and fraudulent emails
• prevent credit card theft and payment fraud by flagging online payments that seem abnormal for the customer, based on past actions, purchase amounts, location, and purchase types

Medical Research and Diagnosis

Researchers and clinicians are starting to use ML to:

• predict pharmaceutical properties of molecular compounds and targets for drug discovery
• use pattern recognition and segmentation techniques on medical images (retinal scans, pathology slides, etc.) to enable faster diagnoses and tracking of disease progression
• use molecular data to classify patients into subtypes at the molecular level to provide more targeted treatments
• identify patients most likely to respond to certain drugs, such as cancer drugs
• diagnose conditions like autism and Alzheimer’s disease at an early stage, enabling better management and treatments
• help physicians predict surgical complications
• improve the likelihood of successful drug repurposing

With ML’s adaptability upon exposure to new data and its decision-making ability, the world has grown bigger and brighter. Currently, ML is highly optimized to perform specific tasks in the industries for which it has received extensive training. In the not-too-distant future, we may see ML applying context learned from one task to new tasks, or tasks in different industries.

Think of the growth!

Steve Pogrebivsky, the president and co-founder of DataGroomr, is an expert in data and content management systems with over 25 years of experience. Steve has founded several technology companies, including MetaVis Technologies, which built tools for Microsoft Office 365, Salesforce, and other cloud-based information systems, and Stelex Corporation, which provided compliance and technical solutions to FDA-regulated organizations. Steve holds a BS in Computer/Electrical Engineering and an MBA in Information Systems from Drexel University. You can follow Steve on Twitter @pogrebs or on LinkedIn (www.linkedin.com/in/pogrebs).