Topics to be covered

Data science is divided into two parts:

a) Data Analysis and Data Visualization

b) Predictive Modeling

A) Data Analysis and Visualization

1: Numpy: Dealing with N-dim array

• Overview
• Creating ndims arrays
• Why do we need arrays?
• Numeric operations using NumPy
• Indexing and slicing
• Some Mathematical functions
• Generate Random array

2: Pandas: Data analysis and Manipulation

• Pandas Overview
• Data Structures
  • Series
  • DataFrame
• Series and DataFrame operations
• Missing Data
• Categorical Data
• Working on DateTime data
• Read data from the different file format
• Merging and Grouping Data
• Many other data operations using Pandas

3: Matplotlib / Seaborn : Data visualization

• Overview
• Scatter plot, line plot, bar plot
• Histogram
• Xlabel, Ylabel, Xticks, Yticks, title
• Marker style,type, size
• Figure and Subplot
• Saving a Figure
• HeatMap,BoxPlot

4: Text analysis using NLTK

• What is NLP?
• NLP libraries
• NLP Applications
• Cleaning text data
• Tokenization
• Removal Stop words
• Stemming and Lemmatization
• part-of-speech(POS) tagging

B) Predictive Modeling using scikit-learn

1: scikit-learn

• Regression
  • Introduction
  • Simple Linear Regression
  • Multiple Linear Regression
  • Polynomial Regression
  • Evaluate Performance of a linear regression model
  • Overfitting and underfitting
  • Regularization
• Logistic Regression
  • Logistic Regression theory
  • Implementing Logistic regression with scikit-learn
  • Logistic Regression Parameters
  • MNIST digit dataset with Logistic Regression
  • Predictive modeling on adult income dataset
• Naive Bayes Classification
  • Theory Naive Bayes Algorithm
  • Features Extraction
    • Countvectorizer
    • TF-IDF
  • Email Spam filtering
  • Sentiment analysis
• Decision Tree and Random Forest
  • The theory behind the decision tree
  • Implementing a decision tree with scikit-learn
  • Decision tree parameters
  • Combining multiple decision trees via Random forest
  • How random forest works..?
• Model Evaluation and Parameter Tuning
  • Cross-validation via K-Fold
  • Tuning hyperparameters via grid search
  • Confusion matrix
  • Recall and Precision
  • ROC and AUC
• Clustering and Dimension Reduction
  • K-means Clustering
  • Elbow method
  • Principal components analysis(PCA)
  • PCA step by step
  • Implementing PCA with scikit-learn

Target Audience

The course can be taken by:

Students: All students who are pursuing professional graduate/post-graduate courses related to computer science or Information Technology.

Teachers/Faculties: All computer science and engineering teachers/faculties.

Professionals: All IT professionals, who wish to acquire new skills or improve their existing skills.

Test & Evaluation

1. During the program, the participants will have to take all the assignments given to them for better learning.

2. At the end of the program, a final assessment will be conducted.

Certification

1. All successful participants will be provided with a certificate of completion.

2. Students who do not complete the course / leave it midway will not be awarded any certificate.