2017 was undeniably the year artificial intelligence (AI) and machine learning (ML) became mainstream topics in the enterprise.
Whether it’s a robot performing a backflip or self-driving Uber vehicles roaming Pittsburgh, organizations in many sectors have started looking for ways to use AI and ML to transform their operations. And this is just the beginning: IDC predicts that spending in this area will exceed $50 billion by 2021.
We know AI and ML will change the way we work in the next several years, but amid all the hype, let’s take a deeper look at exactly what ML means for businesses everywhere today.
Machine Learning in Data Analytics
The most mature deployments of ML are in the areas of data analysis and processing, where the technology is capable of exceeding levels of human accuracy in some domains.
Major tech companies like Amazon, Google, IBM and Microsoft, as well as a number of smaller vendors, have exposed powerful ML algorithms via public APIs, which can be used to process massive amounts of data and extract value.
What does this mean for your business? There is an opportunity to use ML to tackle one of the most daunting challenges businesses face: the accelerating volume of data. App Developer Magazine reports that more data was created in 2017 than in the previous 5,000 years of humanity, and Gartner estimates that nearly 80 percent of this data is unstructured, meaning it lives in formats like PDFs, images and videos. The process of managing, structuring and deriving value from this growing cache of information is challenging, time-consuming and expensive — if it happens at all.
Where the Data Comes From
Business process digitization has enabled new ways of working, which in turn have led to the production of all of this unstructured data. Here are a few key contributing factors:
- The elimination of paper: More and more businesses are scanning their paper records and moving them to the cloud.
- The ease of content creation: Modern software applications make it incredibly easy for users to create new data digitally; essentially, every business user is now a content creator.
- The explosion of mobile capture: Because millions of smartphones are in use today, just about everyone can easily capture and share data in the form of photos, videos and more.
Using Machine Learning to Add Structure
With most of the technologies businesses use to manage information today, the more data you have, the harder it is to make sense of it all. But with the cloud and ML, the opposite is true. Using machine learning, businesses can add structure to their content and automate processes around their information. This would save hours of manual review and data entry, which would translate to real cost savings and optimization for your business.
Here are three domains in which ML solutions are relatively advanced, with examples of the types of tasks ML-driven systems can handle:
- Image and video: object and landmark detection, content moderation, facial (and celebrity) recognition, optical character recognition, handwriting analysis, bar-code recognition, scene change detection and motion detection.
- Audio (speech): transcriptions, keyword extraction, profanity filtering, speaker recognition, emotion analysis and demographic analysis.
- Text: entity recognition, topic and concept recognition, sentiment and tone analysis, translation and language detection.
These capabilities can be applied in a number of different ways on the job to improve employee productivity, accelerate manual or inefficient processes and even mitigate risk.
For example, company training materials, like videos and audio files, could be parsed into useful segments so new employees can jump to the parts of these files most relevant to them. Images could be automatically categorized and labeled based on their content to make it easier to search them for marketing assets for an ad campaign. Audio recordings of customer support calls could be transcribed and analyzed for sentiment and tone to improve the performance of agents. And government identification documents like driver’s licenses could be verified and key data could be extracted to make the employee onboarding process more efficient.
Learning Opportunities
Webinar
Mar
29
The CX Commute vs. the CX Joy Ride
Is your CX traffic holding up your customer's journey?
Webinar
Mar
29
Is Your Microsoft 365 Data Recoverable? How to Survive the IT Hotseat When Data is Lost
See why MIcrosoft recommends third party backup
Webinar
Mar
30
An Optimized Customer Contact Strategy Combines Transparency and Intelligence: The State of Outbound Communications
Learn from the Neustar-commissioned Forrester Consulting study on challenges and solutions for contact centers
Webinar
Apr
5
Adopt a Modern Digital-First Marketing Strategy
Learn how to use a composable DXP to your advantage
Assess Your Business Needs
To get the most out of machine learning technology, you must first identify use cases that can benefit from the available tools. For example, you might decide to use video recognition to analyze security footage to automatically detect anomalies and recognize faces. Or you might want to use a text extraction tool to automate the process of pulling information from a driver’s license.
You could even string together multiple services to gain larger-scale benefits. For example, you could analyze customer support interactions at scale by transcribing the audio files of support calls and then using a sentiment analysis tool to analyze the conversations.
To determine where ML capabilities can do the most for your company, evaluate the areas of your business that are process-heavy, especially those made up of manual, repeatable tasks involving files. Then verify which elements of those processes can be automated. For example, to make the process of analyzing security footage more efficient, you could use ML to detect exactly when each face appears.
Evaluate Your Vendor Options
Once you have identified areas where ML can benefit your company, you will need to find and evaluate the offerings of technology vendors to determine which ones can help with your use cases. Beyond just the quality and performance of the systems and the way the vendors will act as partners, you will need to determine exactly how you would implement the technologies. Some vendors offer packaged solutions that are focused on applying ML to very specific problems, while others offer public APIs that can be integrated into your existing systems and processes.
You will also need to be clear about your regulatory and security needs, because vendors in the emerging ML market will have varying profiles when it comes to security and compliance. It is important to ask prospective partners how they will store, secure and analyze your information, and what steps they will take to keep your data separate from your competitors’ data.
AI and ML Will Change the Way We Work
AI and ML technologies are still evolving. However, it’s clear with the new systems powered by ML, the way we work with and manage content will change tremendously in the next five years. Content will be automatically surfaced when you need it, you’ll be able to ask questions of your data instead of just reading it. Content will be auto-created, auto-edited, auto-quarantined and more.
It may be a while before we realize the true promise of AI and ML, but with the right approach, businesses can start implementing AI and ML strategies by taking advantage of the innovations on the market today.
Learn how you can join our contributor community.
