This site uses different types of cookies, including analytics and functional cookies (its own and from other sites). To change your cookie settings or find out more, click here. If you continue browsing our website, you accept these cookies.
Tangent Works recently joined the Alteryx Alliances program as a technology partner. We’re excited to have Dirk Michiels, CEO of Tangent Works, share how using Tangent Information Modeler (TIM) and Alteryx can allow you to experience The Thrill of Solving.
Alteryx + Tangent Works bring a unique combination of technology that enables those who work with time series data to easily benefit from machine learning for predictive and prescriptive scenarios.
The Tangent Works TIM solution is an easy-to-use, fast, and accurate solution for time series model generation, prediction/forecasting, and anomaly detection in an explainable AI way.
Alteryx Designer fully integrates with Tangent Works TIM. Benefit from the world-famous Alteryx functionality that allows business users, citizen data scientists, and data scientists to work magic with data.
Tangent Works’ goal is to make machine learning easy to apply and available to users without the complexity of traditional auto ML model selection, time-consuming data science, and tedious manual feature engineering.
Essential to citizen data science is augmented machine learning/analytics with automatic feature engineering, model generation, and deployment whilst offering understandable models.
Dealing with time-series data machine learning is hard but has enormous potential to deliver deeper insights and faster decisions. TIM gives businesses value rather than keeping their talent stuck on complicated machine learning solutions.
In this blog post, we’ll give you a brief introduction of features, benefits, and Tangent Works’ vision.
Time series data is everywhere, in many different industries. The Alteryx vision is to create solutions that allow users to experience The Thrill of Solving. That’s why we are so excited to make TIM available within the Alteryx Community.
If you want to see how it works in Alteryx, check out the Data Science blog post. If you want to better understand Tangent Works’ vision, read on.
InstantML + RTInstantML: The Tangent Works Secret Sauce
Many companies offer AutoML capabilities to support the model generation and feature engineering process, allowing businesses to build, tune, compare, and test multiple models and deploy the selected model as an application programming interface (API).
Although AutoML offers a step forward, it remains time-consuming and exclusive to experts. Tangent Works surpasses AutoML with its InstantML technology offered through Tangent Information Modeler (TIM), which generates models in seconds, delivering real augmented machine learning.
TIM creates models in seconds based on target and predictor series with just one pass through the data. These models are made available through a single API, thus saving organizations the hassle of creating separate APIs for each of the models. Tangent Works has further enhanced TIM’s capabilities with real-time, instant machine learning technology (RTInstantML) that provides organizations with optimally trained models for the data available at any time with immediate forecasting or anomaly detection. RTInstantML not only simplifies the model selection and automates feature engineering, but also vastly shortens the lengthy process of developing the right model at the right moment, which is crucial for time series problems. The dynamics in these problems change when new data is collected, and new data availability can also change from one minute to another. TIM’s ability to generate a new model on the spot without any specific settings enables immediate forecasting and anomaly detection.
Years of research have led to the creation of the TIM engine, based on a unique blend of various advanced technologies, including information criteria/geometry — an interdisciplinary field that uses differential geometry techniques to study probability theory and statistics.
TIM on Alteryx
TIM Forecasting and TIM Anomaly Detection Tools are available for Alteryx Designer through .yxi installation files. It is recommended to use Alteryx Designer version 2019.2 or higher. The TIM integration is compliant to the Alteryx rules for integration. TIM can be accessed from the Alteryx Gallery. This streamlines the process by delivering a repeatable workflow for self-service data analytics, leading to deeper insights in hours, not weeks, using Alteryx Designer. Alteryx Designer empowers data analysts further by now offering advanced time series machine learning capabilities.
The Tangent Works TIM Solution
The Tangent Work solution consists of three major parts:
TIM Engine delivers all the machine learning functionalities, ranging from time series model generation to forecasting/prediction, anomaly detection in InstantML, as well as one called RTInstantML with integrated model generation and forecasting /prediction.
TIM Engine can be used as a TIM cloud service, in your own cloud, on premise, or on the edge.
TIM Studio is designed to serve as a machine learning development and operations (MLOps) platform for time series modeling.
TIM Studio can be used by several personas, ranging from business user and citizen data scientist to data scientist. It gives insight into the catalogue and versions of models; forecasts available; and covers data preparation; business process orchestration, including continuous data integration; and more.
With the “consume” strategy, Tangent Works’ vision remains the same: make it easy for TIM to be used by a variety of tools and platforms. TIM clients are available not only for Alteryx but also for QLIK, Tableau, Denodo, Cloudera, and Microsoft or Microsoft Azure products, including Azure Data Lake, Azure Databricks, Azure SQL Data Warehouse, Microsoft SQL Server, Microsoft Excel, and Microsoft Power BI.
Business Value Use case
Machine learning is a buzzword right now. But it’s not just about the machine learning technology; it’s about the value it brings. Many of its concepts were developed as early as the 1940s and 50s. After the AI winter of the 70s, backpropagation — a mechanism to efficiently train neural networks — gave rise to various new use cases of ML. Our vision is to create business value with machine learning. Potential business use cases for time series data forecasting and anomaly detection are endless.
That’s why Tangent Works loves the Alteryx Community. It’s all about The Thrill of Solving, giving people the tools to solve complex business problems.
Tangent Works’ Vision and Advancements in Data Science
Tangent Works delivers an augmented data science solution for time series data. It offers InstantML and RTInstantML, which go beyond the typical auto ML solutions. This makes implementing predictive and prescriptive analytics much easier.
The continuous development and enrichment of our InstantML technology is high on our priority list. Various other features are at the forefront of our vision and strategy. Here’s a quick overview.
AI Service – In the Cloud
TIM cloud services are hosted services that allow applications, business productivity tools, and platforms to incorporate the advantages of time series machine learning.
The TIM Engine functionalities are available via easy API access. Using the higher-level building blocks makes the introduction of machine learning in various applications significantly easier.
Tangent Works can be deployed in the public cloud, making it extremely easy to use. But it can also be deployed on your own cloud subscription, on premise or on the edge.
For the Alteryx user, TIM Cloud services can be up and running in seconds.
The concept of containerization allows a TIM Engine to reside anywhere in the IT infrastructure. This is very powerful, versatile, and provides a growth path for supporting more sophisticated architectures. Using an AI services concept significantly reduces the complexity of expert data science and reduces the amount of IT effort required to bring machine learning to the business.
For any data scientist, citizen data scientist, or business user, it is critical to understand and offer transparency for the machine learning models that are generated.
TIM generates a transparent formula that provides organizations with comprehensive insights into the models created along with helping them measure the impact of predictors on target values, automatically and instantly. This allows organizations to build physical reality models that are easy to understand and can be leveraged for forecasting and anomaly detection on time series data — both single and multivariate use cases. Business users can utilize TIM out of the box, while data scientists can create and change dictionaries, developing new features as they see fit.
Edge computing is on the rise. With edge analytics, machine learning modeling is brought closer to the sensors and “Internet of Things” devices. This offers opportunities to deliver “swarm intelligence” based solutions, offering collective behavior based on distributed self-organized systems.
Due to a low-weight footprint and full containerization, the TIM Engine can easily be deployed on the “edge,” bringing not only model building but also forecasting, prediction, and anomaly detection closer to where the data is being generated. TIM can even be deployed on a Raspberry PI. This fully aligns with the market evolution to edge computing, IoT, hybrid cloud, and the shifting balance between edge and cloud computing. Therefore, TIM integrates with a growing number of IOT platforms, such as Azure, IoT central.
For the Alteryx Community, this brings an enormous opportunity to use the richness of the Alteryx environment in a distributed environment and offers many new business solutions.
MLOps, machine learning operationalization ensures versioning, release, activation, monitoring, and overall governance of the ML models. If you are only dealing with a few models, this is no big deal. When you really want to create value with machine learning, chances are that you will be using many different models. TIM Engine can support many different models in an InstantML or RTInstantML model.
Tangent Works TIM studio is a setup studio for business users and citizen data scientists and an engineering workbench for data scientists. It provides insight into the models in catalogue. TIM engine logs telemetry data giving you full insight into the usage of ML in your organization. The usages can also be visualized in your favorite BI analytics and visualization tools.
TIM Studio is accessible from within the Alteryx environment.
TIM delivers the value of machine learning without the complexity. TIM provides business users and citizen data scientists with a solution for making predictions, prescriptions, and finding anomalies and outliers. TIM’s augmented machine learning data science automates fundamental aspects of advanced modeling, such as feature and model selection. TIM provides model explanation and automates aspects of model management.
TIM delivers anomaly (or outlier) detection by detecting the deviation from the norm. TIM engine uses the modeling capability to create a normal behavior model. This is then compared with the real data to detect early notice of potential anomalies. These functionalities allow for predictive maintenance scenarios, process monitoring, validation, and many more. The machine-learning-driven anomaly detection provided by TIM allows organizations to spot inconsistencies from routine conditions and plan for extra attention or even automatic intervention.