ThingWorx Analytics Training: Module 1 Part 2
- December 23, 2022
- 0 replies
- 861 views
This video continues Module 1: ThingWorx Analytics Overview of the ThingWorx Analytics Training videos. It covers some of the functionality of the ThingWorx platform, as well as ThingWorx Analytics capabilities.
This video does not contain any hands-on exercises, and you don't need access to a running ThingWorx Analytics environment.
This video has chapters — use the chapter menu in the video player to jump straight to any section.
Chapter Summaries
0:00 – Process Framework: ThingWorx Analytics Overview & the Three Analytics Questions
This intro maps the end-to-end ThingWorx Analytics process framework as a logical step-by-step progression, framing the three guiding questions: why something happened, what will happen, and what you can do to change the outcome.
0:18 – Descriptive Analytics: Historical Data, Key Factors & Pattern Detection
Explains descriptive analytics as the most basic analytics type, condensing big data into simpler information (similar to Google Analytics) to investigate historical data over a past period, identify key factors, and surface emerging patterns.
1:02 – Predictive Analytics: Forecasting with Statistics, Data Modeling & Mining
Covers predictive analytics, which studies recent and historical data using statistics, data modeling, data mining, and analytics to forecast likely future outcomes with a probability or score rather than a certainty.
1:51 – Prescriptive Analytics: Actionable Plans, Predictive Models & Feedback Systems
Describes prescriptive analytics as a form of predictive analytics that translates forecasts into feasible action plans, requiring a predictive model with an actionable component plus a feedback system that tracks the adjusted outcome of each action taken.
2:47 – Data Pipeline: Contextual Repository, Dimensional Model, Feature Engineering, Signals, Clusters & Profiles
Walks through the ThingWorx Analytics data journey — ingesting varied source data, structuring it in a contextual repository joined by keys like location and time into a dimensional model, feature engineering static and time-series features, and deriving signals, clusters, and profiles to build tactical, optimized action plans.
4:55 – Data Flow: Use Case Scoping, Automatic Metadata Detection, Analytics-Ready View & Champion Model
Details the analytics data flow from exploring raw source data and defining a use case through structuring data, auto-detecting metadata, and constructing the analytics-ready view, then iterating models in the ThingWorx Analytics engine to expose a champion model deployed via a smart connected mashup.
6:43 – IoT Trends: Cheaper, Smaller, Low-Power Sensors Driving Next-Gen Predictive Analytics
Explains how sensors are getting cheaper, smaller, more accurate, and lower-power, expanding into more devices, machines, and people, making advances in the IoT the primary driver of next-generation predictive analytics across many verticals.
7:27 – ThingWorx Analytics 9: Confidence Models, Confusion Matrix Weights, Edge Scoring, Text OpType & Date/Time Support
Summarizes new features in ThingWorx version 9, including confidence-model training, confusion matrix weight parameters to minimize misclassification cost in Boolean models, an edge predictive scoring service, high-availability architecture alignment, ordinal and categorical goals for signals and profiles, validating models on training records, clearer bubble plots, automatic empty-row removal, the new text optype for comments and notes, and expanded standard date/time format support.
8:59 – Module Wrap-Up: ThingWorx Analytics Overview Recap, Upcoming Modules & PTC Contact
Closes Module One: ThingWorx Analytics Overview, previews the upcoming modules in the course, and directs viewers to contact a PTC customer success representative to learn more about using ThingWorx Analytics.