MIT602 · Semester III · Official Syllabus

Data Analytics and Visualization Syllabus

Time series forecasting, machine learning, deep learning, and text analytics with Python.

Unit 1: Data Visualization

Direct plotting with Pandas, advanced charting with Matplotlib, and distribution/bivariate visualization with Seaborn.

  • Direct Plotting: Line Chart, Bar Chart, Pie Chart, Scatter Plot, Box Plot, Histogram
  • Matplotlib: Line, Bar, Histogram, Scatter, Stack Plot, Pie Chart, Heatmap
  • Using text() and annotate() methods to add text in charts
  • Creating multiple charts in same plot
  • Seaborn: Strip Plot, Box Plot, Swarm Plot, Joint Plot

Unit 2: Time Series Forecasting

Time series components, ACF and PACF analysis, univariate forecasting (AR, MA, ARMA, ARIMA, SARIMA, SARIMAX), multivariate forecasting (VARMA, VARMAX), and exponential smoothing.

  • Concept and Components of Time Series
  • Autocorrelation Function (ACF) and Partial Autocorrelation Function (PACF)
  • Correlogram, Plotting ACF and PACF
  • Univariate Forecasting: AR, MA, ARMA, ARIMA, SARIMA, SARIMAX
  • Multivariate Forecasting: VARMA, VARMAX
  • Smoothing: Simple Exponential Smoothing, Holt-Winter's Exponential Smoothing

Unit 3: Data Analytics with Machine Learning

Scales of measurement, feature engineering, EDA, performance metrics, KNN, decision trees, SVM, ensemble methods (XGBoost, LightGBM), PCA, and ICA.

  • Scales of Measurement: Nominal, Ordinal, Interval, Ratio
  • Feature Engineering and Exploratory Data Analysis (EDA)
  • Performance Metrics: MSE, RMSE, MAE, R², Confusion Matrix, Accuracy, Recall, Precision, F1-Score, Specificity
  • Regression and Classification using KNN
  • Decision Tree, Attribute Selection Criteria, ID3 vs C4.5 vs CART
  • Random Forest (Bagging and Boosting)
  • Linear and Non-linear SVM, SVC, SVR
  • Gradient Boosting: XGBoost and LightGBM
  • Dimensionality Reduction: PCA and ICA

Unit 4: Data Analytics with Deep Learning

MLP for regression and classification, gradient descent optimizers (Momentum, RMSProp, Adam), RNNs, LSTM, GRU, Transformers, Autoencoders, and GANs.

  • Regression and Classification using Multilayer Perceptron
  • Gradient Descent Optimizers: Momentum, RMSProp, Adam
  • MLP vs RNN
  • Recurrent Neural Networks (RNN)
  • Vanishing and Exploding Gradient Problem
  • Long Short-Term Memory (LSTM)
  • Gated Recurrent Unit Networks (GRU)
  • Concept of Transformer
  • Auto-Encoder
  • Generative Adversarial Networks

Unit 5: Text Analytics

Text preprocessing, feature representation (one-hot, count, TF-IDF, word embeddings, N-grams, hashing), similarity, POS tagging, entity extraction, topic extraction, spam-ham classification, sentiment analysis, and Transformers in NLP.

  • Tokenization, Lower Casing, Stop Word Removal, Stemming and Lemmatization
  • One Hot Encoding, Count Vectorization, TF-IDF Vectorization
  • Word Embeddings, Generating N-Grams, Hash Vectorization
  • Finding Text Similarity
  • POS Tagging, Entity Extraction, Topic Extraction
  • Spam-Ham Classification
  • Sentiment Analysis
  • Transformer in NLP
  • Programs using NLTK