Authored by Sanveer Osahan, Software Engineer

Machine Learning (ML) has emerged as an exciting and new technology for many industries, however, it can be complex and difficult to understand. The fast growth of ML has created a demand for off-the-shelf applications that are simple to use, without requiring expert knowledge. The off-the-shelf “AutoML” automates the end-to-end process of applying ML to real-world problems and improves efficiency by accelerating research on ML.

Amazon Machine Learning (Amazon ML) is one such player in the market, who are promoting ML development and adoption in cloud-based Auto ML processing models. AWS launched its ML services in 2015. It announced many AutoML capabilities such as Amazon Rekognition, Comprehend and Translate services at re:Invent 2018.

With Amazon ML, you can build and train predictive models, and host applications in a scalable cloud solution. The visualization tools and wizards available through Amazon ML can guide an individual through the process of creating a new machine learning model without learning complex ML algorithms and technology.

We have compiled several use cases and corresponding AWS ML services that are easily utilized for ML processing:

Use Cases	Service Description	Benefits
Customer Analytics: This insight can be used to improve product and services. This analyses customer interactions in support - emails, social media posts, online comments, telephone transcriptions, etc., factors that drive the most positive and negative experiences . Knowledge Management: Organize and categorize documents by topic for easier discovery. Personalize content by recommending other articles related to the same topic.	Amazon Comprehend is a Natural Language Processing (NLP) uses machine learning to easily discover insights and relationships in text.	Amazon Comprehend is fully managed, with no servers to provision, and no machine learning models to build, train, or deploy. Automatically extract key phrases, entities, sentiment, language, syntax, topics and document classifications with confidence scores. No machine learning expertise required. You can simply provide your labels and a small set of examples for each and Comprehend takes care of the rest.
Voice analytics: Discover customer insights and improve contact centre performance. Media & Entertainment: Automate closed captioning and subtitling workflows for greater accessibility. Advertising: Improve user engagement by displaying relevant content and targeted ads to end-users. Search & Compliance: Transcribe audio and video assets into fully searchable archives for compliance monitoring and risk management.	Amazon Transcribe is an advanced machine learning technology that recognizes speech in audio files and transcribes them into text.	Custom vocabulary: Upload a custom vocabulary to improve the accuracy of speech recognition for product names, domain-specific terminology, or names of individuals. Speaker identification: Detect different speakers in your audio with high accuracy and confidence to produce intelligible transcriptions ("who spoke when"). Streaming transcription: Transcribe audio to text in real time. You can send an audio stream and receive a stream of text in real time.
Product Demand Planning: You can forecast the appropriate inventory levels for your various store locations. Export your forecasts in batch in CSV format and import them back into your retail management systems so that you can determine how much inventory to purchase and allocate per store. Financial planning: Accurate financial forecasting such as sales revenue predictions is fundamental to every business success. Amazon Forecast can forecast key financial metrics such as revenue, expenses, and cash flow across multiple time periods and monetary units.	Amazon Forecast builds forecasts using machine learning combining time -series with additional variables.	Amazon Forecast does not require ML experience. It requires historical data, and any additional data that may impact your forecasts. Once the data is provided, Amazon Forecast examines it, and automatically identifies what is meaningful, then it produces a forecasting model capable of making predictions. Forecasts by Amazon Forecasts are up to 50% more accurate due to ML which automatically discovers time series data and other variables like product features and store locations, etc. that affect each other.
Immediate response for public safety and security:Create applications that help find missing persons in social media video content. By recognizing their faces against a database of missing persons that you provide, you can accurately flag matches and speed up a rescue operation. Face-based user verification: Your applications can confirm user identities by comparing their live image with a reference image.	Amazon Rekognition is based on the same proven, highly scalable, deep learning technology developed by Amazon's computer vision scientists to analyse billions of images and videos daily and requires no machine learning expertise to use.	You have to provide an image or video to the Rekognition API, and the service can identify the objects, people, text, scenes, and activities, as well as detect any inappropriate content. Amazon Rekognition makes it easy to add visual analysis features to your application with easy to use APIs that don't require any machine learning expertise.
Build automated document processing workflows:Provide the inputs required to automatically process forms without human intervention. Maintain compliance in document archives:Automatic identification of data types and form label, enabling easy to maintain compliance with information controls. For example, an insurer could use Amazon Textract to feed a workflow that automatically redacts personally identifiable information (PII) for review before archiving claim forms by automatically recognizing the important key-value pairs that require protection.	Amazon Textract take out text and data from scanned documents, this extraction is done automatically.	No code or templates to maintain: Amazon Textract's pre-trained ML models removes the need to write code for data extraction. Since the software already knows millions of documents from virtually every industry, including invoices, receipts, contracts, tax documents, sales orders, enrolment forms, benefit applications, insurance claims, policy documents and more, code maintenance is not needed for every document or form that is received. In addition, changes in page layouts over time are effortless. Extract data quickly & accurately: Amazon Textract makes it easy to quickly and accurately extract data from documents and forms.

 

 

To summarize, Amazon ML helps create meaningful insights from large data sets, generate billions of predictions and deliver them in real-time with a high rate of return. There are no upfront costs for AWS ML since the consumer pays per use only. This also facilitates starting small and scaling the application as the business grows.

Infogain counsels its clients in many industries in their decision between a variety of AWS service options, settings and configurations to deliver value to their business in a very short cycle, and scale per business needs.

We invite you to reach out to us at “digitaltransformations@infogain.com” for any questions on AWS AutoML solutions for your business.