Empower your data to make recommendations, take action preemptively and gain market insights
(ML) is the training of a computer system to make predictions using statistical models based on data, an important underpinning of AI.
(AR & VR) leverage advanced human-machine interfaces that may incorporate both language and machine vision solutions as part of the overall functionality, but are distinct in that they have real-time interaction.
Solutions which rely on images or video data. Use cases include identifying and/or tracking objects in video/imagery, detecting changes over time and facial recognition.
|Amazon Rekognition Image||Amazon Textract||Amazon SageMaker|
Solutions that involve speech or text. Use cases include interpreting speech as text, or transmitting text as speech, as well as parsing text for key terms, sentiment analysis, language translation or other interpretations of text data such as finding key words and discovering topics in a document corpus or recorded conversations.
|Amazon Transcribe||Amazon Translate||Amazon Comprehend|
Solutions that involve making estimates about the unknown or the future.
Common use cases include:
Solutions which need to identify clusters of commonality or normal/anomalous distinctions. Common use cases include things like finding distinct customer segments, or finding similar products in a catalog, or finding anomalous events that stand out such as finding similar customers in order to recommend products to an individual, or detecting in data when conditions have had a departure from normal operating conditions
Solutions which seek to leverage advanced human-machine interfaces, such as chatbots or AR/VR/XR. These may incorporate both language and machine vision solutions as part of the overall functionality, but are distinct in that they have real time-human interaction as a key element
A machine vision HIPAA-Compliant solution synchronizes data in real-time between multiple camera systems and includes a workflow which correctly detects events by evaluating the video stream and categorizing each image state as inside or outside the body. This solution provides real-time, remote access to critical surgical information.
A petrochemical company was filling their tanks too often and making unnecessary trips to tanks that were still 75% full, resulting in costly excess labor. Onica implemented an IoT + AI/ML solution to monitor fill level, predict supply depletion, and prioritize delivery route/schedule to ensure no tank levels fell below 10%. Results: cut operation costs in half while keeping 100% customer satisfaction.
A manufacturer wanted to predict when their equipment was failing so they could order replacements parts before the system failed. This helped them maintain a lean inventory with a streamlined supply chain, reducing staff count, and provided them the ability to notify their customers in advance when a part needed to be replaced.
A major online retailer needed to optimize packaging options and predict potential shipping damage factors. Onica developed a machine learning model to predict transportation cost savings associated with changing packaging types for products in the retailer’s catalog. Using returned item data, Onica developed a predictive model to determine which product combinations likely result in damage when packaged together
A global customer service corporation needed an automated conversational interface on their website capable of interacting with the customer naturally, answering questions around specific products and services. Onica leveraged AWS Lex to develop a chatbot that learned from customer interactions and progressively increased its capabilities through analytics from stored customer sentiments.
Onica created new agent training in VR on how to do insurance appraisals without the need to go out into the field. The virtual training system walks new agents through a comprehensive set of scenarios, where a trainer evaluates how well each trainee does and gives feedback on things that were missed. This solution fast-tracks training and enables new agents to gain years worth of experience in a few weeks.
All businesses have data and questions about that data. Can the answers give you a competitive advantage or make you more operationally efficient and profitable? If the answer is yes, then your business can leverage ML to extract those answers. Learn how Generative Adversarial Networks (GANs) solve problems in ways that are traditionally associated with human cognition and learning.
Spireon is the vehicle intelligence company. They have a connected vehicle platform that powers their automotive and commercial applications and supports over 4-million connected vehicles. Spireon needed a secure and scalable platform that could support existing devices in the field as well as the growing number of new devices added to the platform each month.
Onica worked alongside the team at Spireon to help build a cloud native data ingestion pipeline that enabled them to quickly run ML models to gain market insights.
Let's work on
Expansive AWS Skill Set
400+ AWS Certifications