Session 1 - Empowering cross-domain industries to realize their vision with ML/AI an enabler (Week 1)
- Recognize the need for machine learning in the cross-domain industry business model
- Demonstrating gap between industry expectation and gained knowledge
- Identifying the building blocks of data-driven business strategies
- Demonstrating gap between industry expectation and gained knowledge
- How to interpret client requirements and needs
- Applications of Machine Learning in specific industry domain (examples)
Session 2 - Gear-up for the Journey– 1 (Week 2)
- Data collection and its challenges
- Intuitionfails in high dimension
- Statistical Inferences – Business Case Study
- Industry-wide employed Data Exploration Techniques
- How to identify and mitigate data balance and Imbalance issues
Session 3 - Gear-up for the Journey – 2 (Week 3)
- Regression and its shrinkage
- Top 10 data Mining Algorithms
- Clustering the distribution
- Common Pitfalls in Machine Learning ( Resampling Bias and Variance, Feature Selection Leakages)
Session 4 - Project 1 – IoT (AI-based preventive maintenance solution for automobiles) (Week 4)
- Importance of IoT solutions in the industry
- Real-time sensor data processing through Spark
- Understanding of Automobile parameters, Impact and use cases
- Big Data architecture to host and store sensor data
Session 5 - Playing with real-time sensor data (Week 5)
- Working in data collection adapters/APIs
- Important features extraction from relevant parameters
- Transforming wealth of data to intelligent features
- Inductive Reasoning -Exploratory Analysis
- Correlation Does Not Imply Causation
Session 6 - Learn Many Models, Not Just One (Week 6)
- Data Alone Is Not Enough
- Overfitting/Underfitting Has Many Faces
- Intuition Fails in High Dimensions (PCA)
- Theoretical Guarantees Are Not What They Seem
Session 7 - Art of Model Identification and Evaluation (Week 7)
- Evaluating the models and their results
- Intuition Fails in High Dimensions (PCA)
- Complicated does not equal better
- Selection of best fit based on a business problem
Session 8 - Demonstration of business perspectives derived from model (Week 8)
- Mapping model outcome inferences with preventive KPIs
- Converting inference to business insights
- Creating executive and operational dashboard
Session 9 - Cloud Implementation and Deployment (Week 9)
- Introduction to a cloud-based platform
- Development and deployment of a model
- Challenges of deploying large scale solutions
- Operationalization of overall solutions
Session 10 - Project 2 – Telecom (Anomaly Detection and Forecasting on Mobile Network KPI) (Week 10)
- Telecom domain fundamentals
- Understanding of Telecom network parameters, Impact and use cases
- Big Data architecture to host and store network data (routing, switching, firewall, etc.)
- Real-time network data processing through Spark/Nifi
Session 11 - Playing with real-world network data (Week 11)
- Working in data collection adapters/APIs
- Transforming wealth of data to intelligent features
- Inductive Reasoning - Exploratory Analysis
- Important features extraction from relevant parameters
Session 12 - Learn Many Models, Not Just One (Week 12)
- Data Alone Is Not Enough
- Overfitting/Underfitting Has Many Faces
- Transforming wealth of data to intelligent features
- Selection of best fit
- Intuition Fails in High Dimensions (PCA)
- Theoretical Guarantees Are Not What They Seem
Session 13 - Art of Model Identification and Evaluation (Week 12)
- Evaluating the models and their results
- Complicated does not equal better
- Intuition Fails in High Dimensions (PCA)
- Selection of best fitbased on business problem
Session 14 - Demonstration of business perspectives derived from model (Week 13)
- Mapping model outcome inferences with actual KPI of business problem
- Converting inference to telecom business insight
- Creating executive and operational dashboard
Session 15 - Cloud Implementation and Deployment (Week 14)
- Development and deployment of a model
- Challenges of deploying large scale solutions
- Operationalization of overall solutions
GOT QUESTIONS?
Connect with our learning advisor now
