Data Science Training in BTM Layout

Module 1: Basic Statistics [4 Hours]

• Random Variables and their Probability Distributions.
• Standardized Values.
• Standardized Random Variables.
• The Normal Distribution.
• The Distribution of the Sample Mean.
• Statistical Inference: Estimation.
• Statistical Inference: Decision Making.
• Decision Making: Hypothesis Testing.

Module 2: Linear Regression [6 Hours]

• Meaning
• Relationship with Correlation
• Assumptions of Linear Regression
• Working with categorical and continuous predictors
• Interaction effects
• Interpretation
• Case Study

Module 3: Logistic Regression [10 Hours]

• Why Logistic Regression?
• Logistic Function description
• The logistic Model [ Log Link ]
• Multinomial Logistic Model
• Interpretation of the parametric coefficients
• Case Study
• ROC curve and Gains chart

Module 4: Decision tree and Case Study for Decision tree [ 6 Hours]

• Need for Non-Parametric approach
• Why decision Tree
• Node Classification
• Entropy
• Where to use Decision Tree
• Case Study

Module 5: Unsupervised Classification Algorithms [6 Hours]

• Understanding the working of Kmeans Algorithm
• Cluster Size Optimization vs Definition Optimization,
• Hierarchical and non-hierarchical
• K- medoid and Fuzzy K means
• Case study for clustering

Module 6: Time Series and Case Study for Time Series [ 10 Hours]

• Definition of Time Series
• Time Series Decomposition
• Simple Moving Average Method
• Weighted Moving Average Method
• Single Exponential Smoothing Method
• Double Exponential Smoothing Method
• Triple Exponential Smoothing Method
• AR Models
• MA Models
• ARMA Models
• ARIMA Models
• sARIMA Models
• Case Study
• Using RMSE and MAPE for Model Performance

Module 7: K Nearest Neighbor’s Algorithm for Classification [ 4 hours]

• Lazy Learning techniques
• Distance Matrix
• The Optimum K value
• Data Transformations
• Model Building on Training Data Set
• Model Validation on Testing Data Set
• Evaluation of Model
• Advantages & Disadvantages of KNN Models

Module 8: Naïve Bayes Algorithm for Multi Class Prediction [ 6 hours]

• Bayesian Theorem
• Probabilities - The Prior and Posterior Probabilities
• Conditional and Joint Probabilities Notion
• Traditional Approach - Extract Important Features
• Naive Approach - Independence of Features Assumption
• Advantages & Disadvantages of Naïve Bayes Models
• Case Study

Module 9: Artificial Neural network with case study [10 Hours]

• Understanding Neural Networks
• The mammalian mimic
• The Activation Function
• The Structure of Network
• Back Propagation, Multi-Layer Perceptron, Feed Forward
• The ANN Model
• Case Study
• Training the Model, Testing and Validation

Module 10: Support vector machines with case study [6 Hours]

• Understanding SVM
• Concepts of Linearly separable vs non-separable data
• Hyper Parameters
• Cost and Gamma Functions
• Case study
• Tuning the Model

Module 11: Recommender Systems [6 Hours]

• Usage of Recommender system
• Collaborative and Singular Recommender Systems
• Association Rule mining
• Cosine Similarity
• Case Study