PG Diploma in Data Science

• Introduction to python
• Variables and strings
• Lists, list manipulation and comprehensions, tuples and dictionaries.
• Conditional operators, If-else statements and logical operators.
• Loops-while, for loops, nested loops.
• Indexing, slicing.
• Exploratory data analysis
• Numpy using python

• Pandas introduction and why is it necessary for data science.
• Series and Data frames.
• Importing data through csv and json files.
• Analysing the data with pandas- handling missing data, descriptive statistics, concatenating, merging and string manipulation.

• Statistics in data science and why is it important.
• Population and sample- with a use case example.
• Parameter and statistic- with a use case example.

• univariate and bi-variate analysis.
• measures of central tendencies.
• central limit theorem.
• measures of dispersion.
• skewness and kurtosis.
• Box plot and outlier detection.
• Correlation and covariance.

• What is probability
• Sample space, random variables.
• types of probabilities.
• Dependent and independent events.
• Probability distribution function.
• Binomial, Bernoulli and poisson distributions.
• normal/Gaussian distribution.

• Data types- qualitative and quantitative.
• Data collection techniques.
• Sampling techniques.
• Convenience sampling, random sampling.
• Stratified, systematic and cluster sampling.

• What is data pre-processing?
• What is exploratory data analysis?
• Uni-variate analysis.
• Bi-variate analysis.

DATA VISUALIZATION:

• What is data visualisation and why is It important?
• Plotting- line plot, scatter plot, box plot, bar charts, heat maps, count plots etc.
• Matplotlib
• seaborn
• Advanced matplotlib and seaborn- with practical implementation.
• Advantages and disadvantages of data visualisation.

• Introduction to machine learning.
• Supervised and unsupervised learning.
• Regression and classification, difference between them.

REGRESSION TECHNIQUES- 

• Simple linear regression and multiple linear regression.
• Estimating the coefficients.
• ERROR METRICS – R-squared, adjusted r squared, MSE, RMSE.

REGULARIZATION TECHNIQUES

• Lasso and ridge regression.
• Difference between the above with a use case example.

CLASSIFICATION TECHNIQUES

• Logistic regression
• Logit and sigmoid functions.
• Why logistic regression is a classification technique.
• Difference between regression and classification.
• Evaluation metrics for classification models- Accuracy, precision, recall, f1-score.
• AUC-ROC

TREE-BASED MODELS:

–DECISION TREES: 

• Overview on Decision trees.
• Nodes, root node and terminal nodes.
• Regression trees and classification trees.
• Advantages and disadvantages of tree based models.
• Gini index, entropy, information gain.
• Overfitting and pruning.

NAÏVE BAYES: 

• Principle of naïve bayes classifier.
• Bayes theorem.
• Posterior probability
• Prior probability of class
• Likelihood

• Re-sampling techniques.
• Cross-validation, K-fold cross validation
• Bias-variance
• Bias variance trade off.

Ensemble methods in tree- based models: 

• What is ensemble learning?
• Bootstrap aggregation.
• Bagging and boosting.
• Random forest-What is random forest and how does it work?
• Difference between decision trees and random forest
• BOOSTING TECHNIQUES:
• What is boosting?
• Boosting algorithms- adaboost, Gradient boosting- XGboost.

K-nearest neighbors: 

• K-nearest neighbor algorithm
• Why is it called lazy learners algorithm?
• How does the KNN algorithm work?
• Curse of dimensionality.
• Pro’s and cons of KNN.

SUPPORT VECTOR MACHINE: 

• Maximum marginal classifier.
• Hard margin and soft margin.
• Support vector classifiers.
• Kernel trick.
• Polynomial and radial.
• Hyper parameters svm.

PROJECT: Backorders dataset where in student has to use ML models and predict the risk of backorders.

• What is unsupervised learning?
• Difference between unsupervised and supervised learning.

PRINCIPAL COMPONENT ANALYSIS- 

• Introduction to dimensionality reduction and why Is it needed?
• Principal components
• eigen values, eigen vectors and orthogonality.

K-MEANS CLUSTERING:

• Centroids
• Optimal value of ‘k’ selection.
• Elbow method.
• Use case In real time.

Hierarchical clustering: 

• Divisive and Agglomerative clustering.
• Application of clustering
• Association rules: Market basket analysis
• Apriori algorithm:
• Metrics- support, confidence, lift.
• Frequent itemset.
• Use-case example.

MATERIALS: COLAB notebook with python code and PPT.

• Introduction neural networks, perceptron.
• Activation functions- Relu, sigmoid, softmax, tanh.
• Gradient descent
• Learning rate
• Optimization techniques and functions.
• Back propagation and vanishing gradients.
• Introduction to tensorflow and keras libraries.
• Real time use case examples and practical hands-on.

PROJECT: Custom dataset will be provided along with problem statement from Kaggle.

• How to work with images -introduction.
• Pixels and resolutions.
• Reshaping, resizing, changing to grayscale images.
• Practical hands-on the above topics.
• Convolutional neural networks- pooling and convolution of images concept.
• CNN architecture
• OBJECT DETECTION- region of proposal, YOLO and SSD.

• Overview of NLP.
• Difference between text mining and text processing.
• Pre-processing, tokenization, stop words removal, normalization, stemming and lemmatization.
• Bag of words, continuous bag of words, TF-IDF.
• Language probabilistic models, n-gram model.
• Hands-on NLTK
• WORD EMBEDDINGS- word2vec, glove, Named entity recognition.
• Recurrent neural networks, Bi-directional rnn’s.
• LSTM- Long short term memory
• Hands-on practice.
• Applications of NLP.

PROJECT: Emotion dataset from KAGGLE will be provided along with the problem statement.

MATERIALS: PPT’S and GOOGLE COLAB notebook with code while practicing will be provided for further reference.

MODULE 1: 

• Intro to SQL
• Downloading workbench and configuring, connecting my SQL server
• What is database management system?
• Why is it essential?
• What is relational database management system?
• SQL commands
• What is a table?
• Entity relationship model

MODULE 2:

• MY SQL basics
• Keys
• Clauses in MY SQL
• Create, update and delete.
• Modify the table
• Select and where clause
• Distinct clause
• Order by, group by
• Having clause
• ISNULL and NOTNULL clause
• LIMIT and OFFSET clause

MODULE 3: 

• Operators in sql
• Arithmetic operators
• Wildcard operators
• AND, OR operators in SQL
• Concatenation operator
• MINUS, DIVISION operator in SQL
• NOT operator
• BETWEEN operator in SQL

MODULE 4: 

• SQL FUNCTIONS
• Mathematical functions in SQL
• DATE functions in SQL
• String functions in SQL
• Numeric functions in SQL
• Aggregate functions in SQL

MODULE 5: 

• JOINS in SQL
• Inner join vs outer join
• Sub queries
• Nested queries
• Co-related queries.

MODULE 6: 

• SQL with python
• Above topics will be covered using a dataset in python

(GOOGLE COLAB NOTEBOOK/ JUPYTER WILL BE USED).