Machine Learning for Analytics Team
summary
A no-code AutoML platform embedded in Qlik Sense builds, evaluates and deploys models from uploaded data sources and cloud/RDBMS connections, selecting the best algorithm and showing confusion matrix and feature importance. On a telecom churn demo it classifies 84% non-churn / 16% churn, highlights drivers (plan type, base fee, penalties), supports API deployable predictions and interactive simulations that reduce churn for an account to 2.36% churn by switching plan, and integrates with SageMaker R/Python and Azure ML.
AI-written summary · about
chapters (15)
- 0:00ML for analytics teams
- 0:23Problem statement & personas
- 3:12Gaps solved with Qlik cognitive engine
- 4:36Qlik AutoML platform
- 7:48Demo setup & data overview
- 9:13Building churn prediction model
- 13:26Model evaluation & feature importance
- 15:13Deploying model for predictions
- 16:59Simulating churn reduction actions
- 18:03Integrating predictions into dashboards
- 20:31Prescriptive analytics for decisions
- 21:38Demo summary & use cases
- 22:19Industry & function use cases
- 22:40Qlik solutions for all personas
- 23:22Closing remarks & Q&A
related talks
16mEmpowering Analysts With No-Code Machine Learning
31mAppDev with Machine Learning and AutoML
41mLeveraging Automated Machine Learning to enable anyone to develop Machine Learning solutions
32mExplainable AI Explained for Developers
23mCustom-made games with machine learning and big data
30mBring AutoML to all users across your business
30mBetter data quality with ML & visualization
28mHow to boost data science routines with DataArt’s AutoML Platform?