With all of the attention on machine learning these days, it’s easy to get overwhelmed with all of the incredible opportunities that are out there. From Artificial Intelligence (AI) to Machine Learning (ML), Amazon’s SageMaker product is an amazing place to begin the journey.
The hardest part is always getting started, so let’s talk about a few concrete examples of where AWS tools can help you go from thinking to doing.
Raw Text to Actionable Information
Let’s say you want to build a system to help your customer service and marketing teams better focus on social media comments that go beyond ‘simple complaining.’ Rather than having a set of people reading each entry or building a set of static filters, you could train a classification model to act as your first line of defense.
This would require three basic steps: gather a large set of examples, classify those examples as ‘critical’ or ‘non-critical,’ and then feed that data into a process to build your model. Voila, you’re done! This might be easier said than done, but this is where AWS shines:
- Amazon S3 or even RDS to store individual examples for use
- Amazon SageMaker Ground Truth to label text samples
- Amazon SageMaker’s BlazingText Algorithm implementation for building your models
- Deploy using any one of SageMaker’s outputs:
Smarter Network Security
Amazon’s built-in algorithms have some great starting points to get moving quickly. From a network security perspective, the IP Insights Algorithm plugs into your network-level data and can help you find unusual activity either in real-time or in a batch-processing workflow.
It’s as easy as two steps:
Want to get started now? The IP Insights Tutorial notebook is a great place to start with a step-by-step walkthrough you can run right now in your own account.
Go Shopping and Plug-in AI
The SageMaker Marketplace is a growing storefront full of pre-built models that are ready to use. It’s a great place to start if you’re brand new to AI and ML and want to dip your toe in the water with a solution that’s ready to go.
Need to identify customers that may churn? Want to be able to detect the type of background noise in an audio file? Need to detect faces and blur them from images? As the marketplace grows, so do the capabilities spanning from the tactical to the strategic and beyond.
Your ability to set up, try, and evaluate multiple sets of AI and ML capabilities goes from fast to instantaneous with these pre-built, pre-tuned, ready-to-deploy models. These models provide a complete toolkit for you to experiment with—and learn more rapidly than anywhere else.
Get Out of Your Way!
Whether you have a need to better understand text or images, documents or audio, or even technical level telemetry, there’s a first step for you in SageMaker. The only thing in your way is time, and Amazon’s focus on removing undifferentiated heavy lifting in building AI and ML models can take you from dreaming about the potential of these tools to realizing it.
If you have questions or need help getting started, schedule a consultation or email us at info@1Strategy.com. We’d love to help!
