Today at AWS re:Invent, Amazon Web Services, Inc. (AWS), an Amazon.com company (NASDAQ: AMZN), announced five new artificial intelligence (AI) services designed to put machine learning in the hands of more application developers and end users – with no machine learning experience required. AWS introduced new services that use AI to allow more developers to apply machine learning to create better end user experiences, including new machine learning-powered enterprise search, code reviews and profiling, fraud detection, medical transcription, and human review of AI predictions.
Machine learning continues to grow at a rapid clip, and today there are tens of thousands of customers doing machine learning on AWS (twice as many as the next largest cloud provider), including many customers that opt to use AWS’s fully managed AI Services like Alfresco, Bayer Crop Science, Cerner, CJ Cox Automotive, C-SPAN, Deloitte, Domino’s, Emirates NBD, Fred Hutchinson Cancer Research Center, FICO, FINRA, Gallup, Kelley Blue Book, Kia, Mainichi Newspapers Co, NASA, PricewaterhouseCoopers, White House Historical Association, and Zola. In the past year, AWS has introduced several new fully managed AI Services like Amazon Personalize and Amazon Forecast that allow customers to benefit from the same personalization and forecasting machine learning technology used by Amazon’s consumer business to power its award-winning customer experiences. AWS customers are interested in learning from Amazon’s vast experience using machine learning at scale to improve operations and deliver better customer experiences, without needing to train, tune, and deploy their own custom machine learning models. Today, AWS is announcing five new AI services that build upon Amazon’s rich experience with machine learning, and allow organizations of all sizes across all industries to adopt machine learning in their enterprises – with no machine learning experience required.
Amazon Kendra reinvents enterprise search with machine learning
Despite many attempts over many years, internal search remains a vexing problem for today’s enterprises, and most employees still frequently struggle to find the information they need. Organizations have vast amounts of unstructured text data, much of it incredibly useful if it can be discovered, stored in many formats and spread across different data sources (e.g. Sharepoint, Intranet, Amazon S3, and on-premises file storage systems). Even with common web-based search tools widely available, organizations still find internal search difficult because none of the available tools do a good job indexing across existing data silos, don’t provide natural language queries, and can’t deliver accurate results. When employees have questions, they are required to use keywords that may appear in multiple documents in different contexts, and these searches typically generate long lists of random links that employees then have to sift through to find the information they seek – if they find it at all.
Amazon Kendra reinvents enterprise search by allowing employees to search across multiple silos of data using real questions (not just keywords) and deploys AI technology behind the scenes to deliver the precise answers they seek (instead of a random list of links). Employees can run their searches using natural language (keywords still work, but most users prefer natural language searches). As an example, an employee can ask a specific question like ‘when does the IT help desk open?’, and Amazon Kendra will give them a specific answer like ‘the IT help desk opens at 9:30 AM’, along with links back to the IT ticketing portal and other relevant sites. Customers can use Amazon Kendra across their applications, portals, and wikis. With a few clicks in the AWS Management Console, customers point Amazon Kendra at their various document repositories and the service aggregates petabytes of data to build a centralized index. Kendra helps to ensure that search results adhere to existing document access policies by scanning permissions on documents so that search results only contain documents for which the user has permission to access. Additionally, Amazon Kendra actively retrains its machine learning model on a customer specific basis to improve accuracy using click-through data, user location, and feedback to deliver better answers over time.
Amazon CodeGuru improves software development by using machine learning to provide automated code reviews and helps organizations find their most expensive lines of code
Just like Amazon, AWS customers write a lot of code. Software development is a well understood process. Developers write code, review it, compile the code and deploy the application, measure the performance of the application, and use that data to improve the code. Then, they rinse and repeat. And, yet, all of this process doesn’t matter if the code is incorrect to begin with, which is why teams perform code reviews to check the logic, syntax, and style before new code is added to an existing application code base. Even for a large organization like Amazon, it’s challenging to have enough experienced developers with enough free time to do code reviews given the amount of code that gets written every day. And even the most experienced reviewers miss problems before they get into customer-facing applications, resulting in bugs and performance issues.
Amazon CodeGuru is a new machine learning service that automates code reviews and finds an application’s most expensive lines of code. There are two components of Amazon CodeGuru – code reviews and application profiling. For code reviews, developers commit their code as usual (support for GitHub and CodeCommit exist today, with more repositories coming over time) and add Amazon CodeGuru as one of the code reviewers, with no other changes to the normal process or software to install. Amazon CodeGuru receives a pull request and automatically starts evaluating the code using pre-trained models that have been trained on several decades of code reviews at Amazon and the top ten thousand open-source projects on GitHub. Amazon CodeGuru will review the code changes for quality, and it if discovers an issue, it will add a human-readable comment to the pull request that identifies the line of code, specific issue, and recommended remediation, including example code and links to relevant documentation.
Amazon CodeGuru also contains a machine-learning powered application profiler that helps customers find their most expensive lines of code. To get started, customer install a small, low-profile agent in their application, so Amazon CodeGuru can observe the application run time and profile the application code every five minutes. This code profile includes details on the latency and CPU utilization, linking directly back to specific lines of code. Amazon CodeGuru can help operators find the most expensive line of code in an application, and it produces a flame graph that helps visually identify other lines of code that are creating performance bottlenecks. Amazon’s internal teams have used Amazon CodeGuru to profile code on more than 80,000 applications over the years. From 2017 to 2018, the extensive use of an internal version of Amazon CodeGuru helped the Amazon Prime Day team at Amazon’s consumer business increase its application efficiency, with a 325% increase in CPU utilization, a reduction in the number of instances needed to manage Prime Day, and 39% lower costs overall.
Amazon Fraud Detector delivers automated fraud detection using machine learning
Tens of billions of dollars are lost to fraud every year by organizations around the world. Today, many AWS customers invest in large, expensive fraud management systems. These systems are often based on hand-coded rules that are time consuming, expensive to customize, and difficult to keep up-to-date as fraud patterns change – all of which results in systems that have lower than desired accuracy. This leads organizations to reject good customers as fraudsters, conduct more costly fraud reviews, and miss opportunities to drive down fraud rates. Amazon has been using sophisticated technology including machine learning to detect fraudulent transactions for more than 20 years, and understands it is a constant cat-and-mouse game with fraudulent actors that requires significant resources to build defenses and to keep evolving them. AWS customers have asked if AWS could share its expertise and experience.
Amazon Fraud Detector provides a fully managed service for detecting potential online identity and payment fraud in real time, based on the same technology used by Amazon’s consumer business – with no machine learning experience required. Amazon Fraud Detector uses historical data of both fraudulent and legitimate transactions to build, train, and deploy machine learning models that provide real-time, low-latency fraud risk predictions. To get started, customers upload transaction data to Amazon Simple Storage Service (S3) to customize the model’s training. Customers only need to provide the email address and IP address associated with a transaction, and can optionally add other data (e.g. billing address, or phone number). Based upon the type of fraud customers want to predict (new account or online payment fraud), Amazon Fraud Detector will pre-process the data, select an algorithm, and train a model – using the decades of experience running fraud detection risk analysis at scale at Amazon. Amazon Fraud Detector also uses machine learning-based data detectors that were trained on data from Amazon. These data detectors help identify patterns commonly associated with fraudulent activity at Amazon (e.g. anomalous email naming conventions) to help boost the accuracy of the trained model even if the number of fraudulent examples provided by a customer to Amazon Fraud Detector are low. Amazon Fraud Detector trains and deploys a model to a fully managed, private application programming interface (API) end point. Customers can send new activity (e.g. signups or new purchases) to the API and receive a fraud report, which includes a fraud risk score. Based on this report, the application can determine the right course of action (e.g. accept a purchase, or pass it to a human for review). With Amazon Fraud Detector, customers can detect fraud quicker, easier, and more accurately.
Amazon Transcribe Medical uses machine learning to transcribe medical speech, allowing health care providers to focus on patient care
Today, physicians are required to conduct detailed data entry into electronic health record (EHR) systems as part of their everyday duties. However, the solutions that help them accurately record and document patient encounters are sub-optimal. At many hospitals, physicians must dictate medical notes into recorders and then submit those voice files to third-party manual transcription services, which is expensive and can take as many as three business days, delaying documentation workflows overall. Another option is to leverage existing front-end dictation software, but existing tools are limited and physicians still end up spending several hours on clinical documentation every day. A third option is for healthcare providers to hire human scribes to assist physicians with notetaking during patient encounters, but human scribes can be unsettling to patients, physicians often mention that their output is imperfect, and medical organizations struggle to schedule and coordinate scribes at scale. Existing solutions fall short, both in terms of enhancing clinical documentation efficiency and enabling better patient care.
Amazon Transcribe Medical solves these problems by using machine learning technology to automatically transcribe natural medical speech. Clinical documentation applications built on top of Amazon Transcribe Medical’s speech-to-text capabilities produce accurate and affordable transcripts. Amazon Transcribe Medical consists of multiple machine learning models that have been trained on tens of thousands of hours of medical speech to deliver accurate, machine learning-powered medical transcription. Transcripts are generated in real time, eliminating the multi-day turnarounds. Amazon Transcribe Medical can help physicians automatically transcribe conversations during the patient encounter without the distraction of manual notetaking, allowing health care providers to focus on patient care. Physicians can speak naturally, and Amazon Transcribe Medical uses built-in automatic punctuation to overcome the limitations of existing transcription software. For healthcare providers, voice solutions built on Amazon Transcribe Medical are scalable to thousands of potential medical centers, removing the operational pain of managing and coordinating temporary scribes. Amazon Transcribe Medical is HIPAA eligible, and offers an easy-to-use API that can integrate with voice-enabled applications and any device with a microphone. Text output from Amazon Transcribe Medical can also be used by other AWS services, such as Amazon Comprehend Medical, a natural language processing service, for downstream data analysis before final entry into EHR systems.
Amazon Augmented Artificial Intelligence (A2I) allows developers to validate machine learning predictions with human reviewers
Machine learning can provide highly accurate predictions for a variety of use cases, including identifying objects in images, extracting text from scanned documents, or transcribing and understanding spoken language. In each case, machine learning models provide a prediction and also a confidence score that expresses how certain the model is in its prediction. The higher the confidence number, the more the result can be trusted. For many use cases, when developers receive a high confidence result, they can trust that the results are likely to be accurate, and they can process them automatically (e.g. automatically moderating user-generated content on a social network, or adding subtitles to a video). However, in situations where confidence is lower than desired, due to some ambiguity in the prediction result, results may require a human review to resolve this ambiguity. This interplay between machine learning and human reviewer is critical to the success of machine learning systems, but human reviews are challenging and expensive to build and operate at scale, often involving multiple workflow steps, custom software to manage human review tasks and results, and recruiting and managing large groups of reviewers. As a result, developers sometimes spend more time managing the human review process than building their intended application, or they have to forego having a human review, which leads to less confidence and utility in many predictions.
Amazon Augmented Artificial Intelligence (A2I) is a new service that makes it easier to build and manage human reviews for machine learning applications. Amazon A2I provides pre-built human review workflows for common machine learning tasks (e.g. object detection in images, transcription of speech, and content moderation) that allow machine learning predictions from Amazon Rekognition and Amazon Textract to be human-reviewed more easily. Developers choose a confidence threshold for their specific application and all predictions with a confidence score below the threshold are automatically sent to human reviewers for validation. Developers can choose to have their reviews performed by Amazon Mechanical Turk’s 500,000 global workers, third-party organizations with pre-authorized workers (including Startek, iVision, CapeStart Inc., Cogito, and iMerit), or their own private reviewers. The results are stored in Amazon S3, and developers receive a notification when review is complete so they can take action based on the trusted results from human reviewers. Amazon A2I brings human review to all developers, removing the undifferentiated heavy lifting associated with building and managing custom reviewing pipelines or recruiting large numbers of human reviewers.
“Companies across various industry segments tell us that they want to leverage Amazon’s extensive experience with machine learning to address some of the common challenges they face as enterprises on an on-going basis. These challenges include internal search, helping software developers write better code, identifying fraudulent transactions, and improving the overall quality of all machine learning systems,” said Swami Sivasubramanian, Vice President, Amazon Machine Learning, AWS. “With decades of experience in building machine learning systems, Amazon has created internal systems that successfully address such challenges, and today’s launches are the next iteration of the same customer obsession that spurred the development of these systems. With these launches, we are excited to make these machine learning-powered capabilities available to enterprise users without requiring any machine learning expertise.”
3M is a multinational corporation and a leading manufacturer of products including abrasives, chemicals and advanced materials, films, filtration, adhesives, and more. 3M applies science in collaborative ways to improve lives, daily. “Research and development is the heartbeat of 3M, and we invest deeply in the science that makes us strong. When our material scientists lead new research, they need access past research that may be relevant. This information is often buried in our patents and expansive knowledge repositories,” said David Frazee, Technical Director, 3M Corporate Research Systems Lab. “Finding the right information is often exhausting, time consuming, and sometimes incomplete. With Amazon Kendra, our scientists find the information they need quickly and accurately using natural language queries. With Kendra, our engineers and researchers are enthusiastic about the ability to quickly find information which will enable them to innovate faster, collaborate more effectively, and accelerate the ongoing stream of unique products for our customers.”
Workgrid Software, a wholly owned subsidiary of Liberty Mutual, delivers software solutions in an employee experience platform to make work more connected, efficient, and productive. “One of our core offerings is the Workgrid Chatbot which gives employees the ability to get quick answers to frequent queries and automate tasks using a friendly, natural language interface. One key part of any enterprise chatbot is the ability to answer the myriad of questions that come from employees, which is why Workgrid offers a self-service Q&A builder that requires no programming language for content authors to train the chatbot to respond to employees’ questions. In addition to this curated content, we want to offer a way for the Workgrid Chatbot to easily extract knowledge from the vast quantities of documents (e.g., PDF documents) that exist across the enterprise,” said Gillian McCann, Head of Cloud Engineering & AI, Workgrid Software. “With Amazon Kendra, we’re excited about the possibility of our customers getting the answers they need quickly and efficiently. Amazon Kendra makes it possible to extract answers directly from unstructured data across multiple repositories and has the potential to fast track our delivery of accurate, better-than-before answers to our customers. We’re excited about exploring the combination of Amazon Kendra’s intelligent search with the conversational context and task automation that we provide to create a powerful employee experience.”
British Broadcasting Corporation is a world leader within the broadcasting industry. They bring the sights and sounds of BBC across the world. “As a global media organization, we manage petabytes of video and run a live operation – 24 hours a day,” said Matthew Postgate, Chief Technology and Product Officer, BBC. “Amazon CodeGuru, along with other dev tools that our team uses, will help ensure we continue to provide our audience with robust, reliable services – and spot any issues before they happen. It will also help us gain insights into how services interact with the AWS Platform, allowing teams to refactor and optimize their code to give people the service they expect from the BBC.”
Apptio SaaS solutions help organizations make smart decisions as they analyze, plan, and optimize investments to transform their IT operating model. “Providing highly available, bug-free services to our customers is critical for our success,” said Scott Chancellor, Chief Product Officer, Apptio. “We’ve been looking for tools to transform our organization to more proactively detect issues at various stages of our app development lifecycle, increase our development velocity, and spend less time fixing difficult issues such as concurrency, resource leaks, and performance bottlenecks. We tried Amazon CodeGuru and found that it provided recommendations to fix these issues proactively during the early development stages. Also, by pointing to the areas of the code that slow down our service, we reduced our time to resolution for performance-related defects. These improvements will help us provide a better experience for all our customers.”
SmugMug+Flickr is the world’s largest and most influential photographer-centric platform. “SmugMug & Flickr together are the world’s largest and most influential photographer-centric platform, built for professional photographers and photography enthusiasts to showcase their work while enjoying the work of others. Since day one, SmugMug’s passion has been finding ways to empower photographers to tell the stories they want to tell, how they want to tell them. When you operate at our scale, performance is top priority for image processing, classification, and search,” says Don MacAskill, CEO and Chief Geek of SmugMug+Flickr. “Amazon CodeGuru’s live profiling helps to troubleshoot and identify inefficient sections of our services, in particular the expensive lines of code in applications, that slows them down and it aids us with recommendations on how to change and optimize them. With CodeGuru’s recommendations, we are able to restructure our code, make it highly maintainable, and improve the performance of our services.”
Charles Schwab is a leading investment services firm. “Detecting fraudulent activity online is a never-ending challenge as bad actors continually create new attack vectors. Our mission is to stay ahead of the bad actors, protecting our clients,” said Kara H. Suro, Vice President, Fraud Surveillance & Investigations, Charles Schwab & Co., Inc. “We are excited about the introduction of Amazon Fraud Detector. It will make it quicker and easier to build machine learning tools to spot fraudulent activity; we anticipate a significant increase in our fraud prevention rate. Once implemented, Fraud Detector will be instrumental in identifying fraud patterns from our historical data as well as using Amazon’s experience detecting fraud.”
Vacasa is the largest full-service vacation rental management company in North America, with more than 23,000 vacation homes in 17 countries serving over 2 million guests per year. “Since our founding, we have used technology to enable our local teams to focus on caring for homes and guests, while maximizing revenue for vacation home owners,” said Eric Breon, Founder and CEO, Vacasa. “We’re excited about the introduction of Amazon Fraud Detector because it means we can more easily use advanced machine learning techniques to accurately detect fraudulent reservations. Protecting our ‘front door’ from potential harm enables us to focus on making the vacation rental experience seamless and worry-free.”
Cerner is a leading provider of health information technology solutions, services, and devices. “Extreme accuracy in clinical documentation is critical to workflows and overall caregiver satisfaction,” said Jacob Geers, Solutions Strategist, Cerner Corporation. “By leveraging Amazon Transcribe Medical’s transcription API, Cerner is in initial development of a digital voice scribe that automatically listens to clinician-patient interactions and unobtrusively captures the dialogue in text form. From there, our solution will intelligently translate the concepts for entry into the codified component in the Cerner EHR system.”
Suki is an AI-powered, voice-enabled digital assistant that lifts the administrative burden from doctors. “Clinical documentation is all about medical data workflows, and helping clinicians capture notes more efficiently is key,” said Punit Soni, CEO, Suki AI, Inc. “We can easily integrate our clinical digital assistant with Amazon Transcribe Medical, allowing physicians to dictate medical notes, thus cutting energy spent on clinical documentation by 76% on average. Their time is better spent focusing on caring for patients, rather than doing data entry work.”
SoundLines, part of the HealthChannels family of companies, connects intuitive AI technology with clinical expertise. Data is harnessed to provide more personalized care, while improving efficiencies for providers and Care Team Assistants. Through a winning combination of data-driven analytics and advanced ideation, and with nearly two decades of experience, HealthChannels has supported millions of patient encounters. “SoundLines supports HealthChannels’ skilled labor force at the point of care by incorporating meaningful and impact focused technology such as Amazon’s medically accurate speech-to-text solution, Amazon Transcribe Medical, into the workflows of our network of Care Team Assistants, Medical Scribes, and Navigators. This allows the clinicians we serve to focus on patients instead of spending extra cycles on manual note taking,” said Vadim Khazan, President of Technology, SoundLines. “Amazon Transcribe Medical easily integrates into the HealthChannels platform, feeding its transcripts directly into downstream analytics. For the 3,500 healthcare partners relying on our care team optimization strategies for the past 15 years, we’ve significantly decreased the time and effort required to get to insightful data.”
Amgen is one of the world’s leading biotechnology companies and is a values-based company, deeply rooted in science and innovation to transform new ideas and discoveries into medicines for patients with serious illnesses. “In pharmacovigilance, we want to accurately review recorded calls from patients or health care providers, in order to identify any reported potential side effects associated with medicinal products,” said George Seegan, Senior Data Scientist of Observational Research and Digital Health, Amgen. “Amazon Transcribe Medical produces text transcripts from recorded calls that allow us to extract meaningful insights about medicines and any reported side effects. In this way, we can quickly detect, collect, assess, report, and monitor adverse effects to the benefit of patients globally.”
As America’s Un-carrier, T-Mobile US, Inc. is redefining the way consumers and businesses buy wireless services through leading product and service innovation. “At T-Mobile we measure success through customer happiness – and as the Un-carrier we know customers are happiest when they feel like we understand and anticipate their needs and directly solve their pain points,” said Cody Sanford, Executive Vice President and CIO, T-Mobile. “Our Team of Experts customer service model – or TEX – is all about creating personal connections and uses cutting edge tools like A2I to set our teams up for success. Yes – machine learning leads to deeper, engaged relationships! Access to real-time contextual information, such as account details and available discounts, empowers our team to be able to make on-the-spot decisions on behalf of customers when they’re having a real, live conversation with them… a total win-win!”
VidMob is a Marketing Creative Platform that provides an end-to-end technology solution for all of a brand’s creative needs. Its one integrated platform combining first-of-a-kind creative analytics with best-in-class creative production to transform marketing effectiveness. “Vidmob utilizes machine learning to analyze every aspect of a video including people, objects, and messages to help brands understand creative performance and build better creative. However, for dimensions not covered by existing machine learning models, it can be challenging to review the creative from the petabyte-scale data we analyze every day,” said Joline McGoldrick, Senior Vice President, Data and Insights, VidMob. “With our current workforce of highly trained creative evaluators, using A2I we can more quickly optimize and fine-tune our predictive models. This efficiency exposes us to a large sample of reviewers and increases the speed of models to market by 3x.”