You are on page 1of 4

Data Science content

Introduction to Data Science:

 What is Data science, Data Analysis and Machine learning?


 Life cycle of Data science
 Technologies and Algorithms Used in Data science
 Roles and Opportunities of Data Scientists

Introduction to Statistics and Bayesian Theory:


 Basic understanding of linear algebra,matrices,vector
 What is Statistics?
 Descriptive Statistics
 Central tendency
 Dispersion Measures
 Bayes Decision Theorem
 Bayesian Classification for normal Distributions
o Gaussian Probability Density function
o Bayesian Classifier for Normal Distributed Classes
 Estimation of Unknown Probability Density function
o Maximum likelihood parameter Estimation
o Maximum a posteriori probability Estimation
 Non parametric Estimation

Modules used in machine Learning:


 Numpy:
 Linear algebra (creating and accessing elements of vectors, matrices,
mathematical operations with arrays, arange function).
 Pandas:
 Creating data frames, slicing, removing unwanted data, sorting, joining, merging.

 Matplotlib:
 Scatter, bar, pie, histogram, subplots.
 Seaborn:
 bar, joint plots, regression plots, box plot, heat map, pair plot .
 Scikit-learn:
 Sklearn module for using algorithms.

Machine learning Algorithms:


 Supervised learning
o Regression
o Classification
 Unsupervised learning
o Clustering
 Reinforcement learning
Supervised learning Algorithms:
1. Regression
 Simple Linear Regression
 Multiple Linear Regression
 Logistic Regression
 Regularization
 Model building
 Model validation
 Model Performance
 Case Study
2. Naïve bays
 Gaussian Naïve bays
 Multinomial Naïve bays
 Bernoulli Naïve bays
 Model building
 Model validation
 Confusion matrix
 Classification Report
 Case Study

3. K-Nearest Neighbors Classifier


 Euclidean Distance
 Model building
 Model validation
 Confusion matrix
 Classification Report
 Case Study
4. Decision Tree Classifier
 What is Decision Tree
 How to build Decision Tree
 Model building
 Model validation
 Confusion matrix
 Classification Report
 Case Study
5. Ensemble methods
 Bagging
 Boosting
6. Random Forest Classifier
 What is Random Forest
 Model building
 Model validation
 Confusion matrix
 Classification Report
 Case Study
7. Support vector machines
 Linear SVM
 Nonlinear SVM
 Large margin Classification
 Kernel Trick
 Model building
 Model validation
 Confusion matrix
 Classification Report
 Case Study

Unsupervised Learning Algorithms


8. Clustering
 What is clustering?
o K-Means Clustering
o Hierarchical Clustering
9. Dimensionality Reduction:
 Principle component Analysis
 Linear Discriminant Analysis

Deep Learning & Artificial Intelligence

10. Artificial neural networks(ANN)


a. Activation functions
b. Forward propagation
c. Gradient Descent
d. Stochastic Gradient Descent
e. Back Propagation
f. Why Deep Networks
g. Case Study
11. Fundamentals of Deep neural networks
 RBM
 Hyper parameters, parameter recipe for practical DNNs
 Grid and Random search
 Monte Carlo Markov Chain
 Gibbs sampling
 Contrastive Divergence
 DBM
 Case studies
o Introduction to Tensorflow
o Practical hyper parameter choices
12. Auto Encoders
a. Mathematical model
b. Variational bound
c. AE Parameter selection
d. Case study
e. Advanced AE networks: Denoising AE, Variational AE
13. Convolution neural networks(CNN)
 Convolution operations
 Pooling vs. Flattening
 Full Connections
 Softmax vs. Cross Entropy
 Model building
 Case Study

14. Recurrent neural networks(RNN)


 Feed-Forward neural networks
 Recurrent neural networks
 Back Propagation
 Long short term memory
 Case Study

15. Projects
a. Natural Language Processing(NLP)
o What is NLP
o Why NLP is Important
o Model building
o Case Study
 Recommender Systems
o Collaborative filtering
o Content based filtering
o Matrix based methods
o Case studies
 Multimedia
o Video content analysis
o Object recognition on Image net
o Multimodal analysis
o Case studies

You might also like