Amazons Analytics and Machine Learning!
Hello Amazon friends, I hope you have been tuning along with my Amazon services blog series. As this week we will be touching on Amazons Analytics and Machine Learning Products. In my opinion Machine Learning is one of the coolest things you can do with AWS, and the services do not fail to impress. When we look at the history of machine learning and its progress, It is amazing just how far we’ve come. From detecting simple shapes and limited to colors, we can now provide full-scale pictures and get a full analytic mock-up of just what we are looking at. For example the “I am not a robot” security puzzles that we’ve all encountered at one point or another, uses machine learning and analytics to confirm just what is in the picture. While there are multiple platforms for machine learning let’s look at what Amazon has to offer us. Let’s start with Analytics:
AWS Elastic Map Reduce (EMR): Platform for running Amazons’ very own Hadoop framework. You can also run other frameworks such as Apache Spark, Presto, Apache HBase, and Apache Flink. From here EMR can analyze the data in several data stores, including Amazon S3, and Amazon DynamoDb.
Amazon Athena: Is a quick tool to analyze data stored in an S3 bucket, using standard SQL statements.
Amazon Kinesis: Allows us to collect, process, and analyze real-time streaming data.
Amazon QuickSight: Amazons own QuickSight is a business intelligence reporting tool, Similar to Javas BIRT, and is fully managed by Amazon.
These were a few Analytical programs Amazon has to offer, now let us look into machine learning:
AWS DeepLens: This is deep learning enabled video camera, with a deep learning software development kit, that allows us to create advanced vision system applications.
Amazon SageMaker: AWS flagship machine learning product. This allows us to build and train your very own machine learning models, and deploy them to the AWS Cloud, where we can use them as a backend in our applications.
Amazon Rekognition: Provides deep learning-based analysis of video and images.
Amazon Lex: Allows us to build conversational chat-boxes, for frontline customer support, and can be spread through multiple applications.
Amazon Poly: This is an amazing text to natural speech converter.
Amazon Comprehend: Uses deep learning to analyze text for better insight and relationships. We can use this for customer analysis or advanced text searching of documents.
Amazon Translate: Used machine learning to accurately translate to several different languages.
So let’s take a look at an example of what these services can do for us, specifically Amazons Rekognition service. We will start by heading to the AWS console, signing in, opening Rekognition, and let’s try a demo. Here you can upload your photo or choose from a number of examples.
In this case, we have a picture of a city skyline. Amazons analytics does quite an impressive job of mapping out the objects in the picture, you can even click show more, for an extended list. I’ve kept it short in this case as I wanted to show the amazing ability to convert the objects into the picture into a JSON format. Furthermore, we can see it also includes the request of the picture itself. I find this very fascinating and I hope I opened up a whole new world for you. While it doesn’t end here this is just a small example of how far machine learning has come, specifically through Amazon. I hope you enjoyed, please stay tuned for next weeks installment of Security, Identity, and Compliance. Thank You.
