Data Science (DS) Training Course in Montreal

What is Data Science? Lifecycle and ecosystem
Data Science vs Data Analytics vs Data Engineering
End-to-end pipeline: data insight impact
Key concepts: data, information, knowledge, and insight
Roles in a data team: Data Scientist, Analyst, Engineer, ML Engineer
Data-driven decision making & experimentation culture
Business understanding and problem framing
Case studies: Netflix recommendations, Uber surge pricing, Amazon forecasting
Tools & environments overview (Jupyter, Colab, VS Code, cloud notebooks)

What is a database and its role in analytics
Relational vs non-relational databases
Tables, records, keys, and relationships
Database design and normalization
ER diagrams and schema structure
CRUD operations (Create, Read, Update, Delete)
Data constraints and referential integrity
Data warehousing basics
Connecting Python, Tableau, and Power BI to databases

Understanding databases, schemas, and tables
SELECT, WHERE, ORDER BY, GROUP BY, and HAVING clauses
Data extraction and transformation using joins and subqueries
Window functions (RANK, ROW_NUMBER, LAG, LEAD)
Common Table Expressions (CTEs) for complex queries
Aggregations, nested queries, and date/time manipulations
Creating and using views for analytics
SQL query optimization and indexing basics
Real-world analytics use cases and mini-projects

Setting up MySQL and MySQL Workbench
Writing and executing analytical queries
Importing/exporting datasets (CSV, JSON)
Stored procedures, triggers, and views
User roles, permissions, and data security
MySQL indexing and performance tuning
Integrating MySQL with Tableau and Power BI

What is NoSQL and why it emerged
Limitations of relational databases for modern data
Core NoSQL database types (Document-based, Key-Value, Column-family, Graph)
Differences between SQL and NoSQL (schema, scaling, flexibility)
CAP theorem and BASE properties
Data modeling strategies in NoSQL systems
When to choose NoSQL over SQL in analytics workflows
Overview of popular NoSQL databases (MongoDB, Cassandra, Redis, Neo4j)
Connecting NoSQL systems to analytics tools

Introduction to MongoDB and BSON structure
Core components: databases, collections, documents, indexes
Installing and setting up MongoDB & MongoDB Compass
CRUD operations (insertOne, find, updateOne, deleteOne)
Advanced queries and filtering
Aggregation pipelines for analytical transformations
Data modelling and schema design for analytics
Indexing and performance tuning in MongoDB
Integrating MongoDB with Python (PyMongo and Pandas)

Descriptive statistics (mean, median, mode, range, variance, std. deviation)
Data distribution and skewness
Probability basics and sampling techniques
Hypothesis testing (Z-test, T-test, Chi-square, ANOVA)
Confidence intervals and p-values
Correlation vs causation
Regression analysis (simple and multiple regression)
A/B testing and experiment design principles
Business interpretation of statistical results

Introduction to Python and Jupyter Notebooks
Variables, control structures, and functions
Working with files (CSV, JSON, Excel)
Handling APIs and JSON data
Using virtual environments and packages
Automating data workflows
Integrating Python with BI tools

Introduction to NumPy and its role in data analytics
Creating and manipulating NumPy arrays
Indexing, slicing, reshaping, and stacking arrays
Understanding array data types and memory efficiency
Vectorised operations and broadcasting
Mathematical and statistical operations on arrays
Random number generation and reproducibility
Working with multi-dimensional (nD) data
Integration of NumPy with Python and pandas

Introduction to pandas and its importance in analytics
Working with Series and DataFrames
Importing and exporting datasets (CSV, Excel, JSON)
Data inspection and exploration (info(), describe(), etc.)
Data cleaning: handling missing values and duplicates
Filtering, sorting, and conditional selections
Grouping, aggregating, merging, and joining datasets
Feature engineering and date-time operations
Combining pandas with NumPy for advanced analysis

Basics of Matplotlib plotting
Seaborn for advanced statistical visuals
Creating bar, line, scatter, heatmap, and box plots
Customizing charts (titles, labels, grids, colors)
Subplots and multiple visual layouts
Storytelling through visuals
Automating reporting with Python scripts

Machine Learning overview for analysts
Train-test split, scaling, and encoding
Regression models (Linear, Ridge, Lasso)
Classification models (Logistic, KNN, Decision Tree)
Model performance metrics (accuracy, precision, recall, F1-score)
Cross-validation and hyperparameter tuning
Model persistence and pipeline creation

Identifying missing values and imputation techniques
Outlier detection using statistical and visual methods
Data normalization and standardization
Encoding categorical variables (label, one-hot, ordinal encoding)
Dealing with inconsistent data and duplicates
Date-time formatting and time-series alignment
Feature scaling and transformation
Data validation and integrity checks
Maintaining data quality across pipelines

What is EDA and why it’s essential in analytics
EDA process: understanding data → cleaning → exploring → interpreting
Univariate analysis (distributions, histograms, box plots)
Bivariate and multivariate analysis (correlation, scatter, pair plots)
Identifying outliers and anomalies visually and statistically
Data summarization using descriptive statistics
Feature distributions and skewness detection
Handling categorical vs numerical data
Using pandas, Matplotlib, and Seaborn for EDA
Project: Perform full EDA on a real-world dataset (sales, HR, or e-commerce)

Best practices in data visualization and storytelling
Choosing the right visualization type for your data
Data-to-insight storytelling with visuals
Using Power BI, Tableau, or Looker for professional dashboards
Interactive filters, drill-downs, and slicers
Designing for non-technical audiences
Visual perception and color psychology in dashboards
Avoiding misleading or biased visuals
Creating automated, real-time dashboards

Power BI: DAX formulas, data modeling, and relationships
Power Query for data transformation in BI
Tableau: calculated fields, filters, parameters, and blending data sources
Google Data Studio basics and integrations
Dashboard design principles for executives
Data refresh automation and scheduling
Role-based dashboard access and permissions
Embedding BI reports into business workflows
Connecting BI tools to APIs and cloud databases

Tableau interface, workspace, and data connection setup
Data blending vs data joinin
Tableau Prep for ETL (cleaning, merging, reshaping data)
Dimensions, measures, and calculated fields
Table calculations and level of detail (LOD) expressions
Dynamic parameters and user-driven interactivity
Advanced visualizations (heatmaps, treemaps, waterfall, box plots, bullet charts)
Creating multi-source dashboards (Excel + SQL + API data)
Story points for narrative data storytelling
Dashboard actions (filter, highlight, URL, and sheet swapping)
Performance optimization and data extracts vs live connections
Publishing dashboards to Tableau Server and Tableau Cloud

Combining data from multiple workbooks and sources
Using Power Query for ETL (Extract, Transform, Load)
Advanced formulas with dynamic arrays (FILTER, SORT, UNIQUE)
Dashboard interactivity with form controls and buttons
Forecasting using Excel functions (TREND, FORECAST, LINEST)
Creating scenario analysis models and sensitivity analysis
Automating recurring reports with Power Query refresh
Linking Excel with Power BI for hybrid dashboards
Building Excel-based analytics projects end-to-end

Introduction to predictive modeling in analytics
Regression and classification fundamentals
Feature selection and model evaluation
Time-series forecasting (ARIMA, Prophet) basics
Predictive dashboards using Power BI or Python integration
Using AutoML for business forecasting
Interpreting model output for decision-making
Model bias and ethical use of predictions
Case study: Sales forecasting using regression

Introduction to big data concepts and architecture
Data warehousing vs data lakes
Overview of cloud data platforms: BigQuery, Snowflake, Redshift
Writing SQL on large datasets efficiently
ETL pipeline concepts (Extract, Transform, Load)
Data partitioning and optimization
Integrating BI tools with big data sources
Data governance and access management
Real-world use case: analyzing millions of rows with cloud SQL

Identifying and defining KPIs & metrics
Root cause analysis and problem framing
Business experimentation and A/B testing
Decision frameworks: RICE, ICE, Pareto analysis
Communicating insights with data storytelling
Building executive summaries from analytics findings
Aligning insights with business goals
Measuring ROI and performance impact
Turning analytics into actionable strategies

Automating Excel workflows with VBA macros
Using Python scripts to refresh reports automatically
Integrating Google Sheets with APIs
Automating BI dashboards with Power Automate or Zapier
Sending automated email reports
Scheduling jobs with Task Scheduler or Airflow
Building reusable analytics scripts
Version control and collaboration with Git
Minimizing manual effort through smart automation

Marketing analytics: campaign tracking, ROI, funnel metrics
Financial analytics: budgeting, forecasting, variance analysis
HR analytics: employee attrition, engagement, hiring metrics
Operations analytics: process optimization, cost reduction
Sales analytics: pipeline performance, conversion ratios
E-commerce analytics: customer segmentation, basket analysis
Product analytics: feature adoption, retention, churn
Supply chain analytics: demand forecasting and logistics
Healthcare, education, and fintech analytics examples

Introduction to ML workflow (train, validate, test)
Supervised vs Unsupervised learning
Bias-variance tradeoff and model generalization
Model evaluation (cross-validation, metrics)
Feature engineering & selection basics
Regularization (L1, L2) and overfitting prevention
Data preprocessing pipelines (scikit-learn Pipelines)
Model interpretability (feature importance, SHAP values)
ML lifecycle: from idea to deployed model

Linear Regression deep dive and assumptions
Polynomial, Ridge, Lasso, and ElasticNet Regression
Non-linear regression modeling
Regularization techniques and cross-validation
Model evaluation metrics (RMSE, R², MAE, MAPE)
Residual analysis and diagnostics
Feature transformations for regression
Case study: Predicting house prices, sales forecasting

Logistic Regression & decision boundaries
Decision Trees, Random Forests, and Gradient Boosting (XGBoost, LightGBM, CatBoost)
KNN, Naive Bayes, SVM — when to use each
Confusion matrix, precision, recall, F1-score
ROC curves, AUC, and threshold tuning
Cross-validation and hyperparameter optimization (GridSearchCV, Optuna)
Feature importance and model explainability
Handling imbalanced datasets (SMOTE, class weights)
Real-world example: customer churn prediction

Clustering algorithms (K-Means, DBSCAN, Hierarchical)
Dimensionality reduction (PCA, t-SNE, UMAP)
Association rule mining (Apriori, FP-Growth)
Anomaly detection and outlier analysis
Evaluation metrics for unsupervised tasks (silhouette score)
Visualizing high-dimensional data
Real-world use: market segmentation, recommendation systems
Combining clustering with supervised models
Feature extraction using unsupervised methods

Neural network architecture & backpropagation
Activation functions and optimization algorithms
TensorFlow and PyTorch essentials
CNNs for image classification
RNNs, LSTMs, GRUs for sequential data
Transfer learning and pre-trained models
Batch normalization, dropout, and regularization
Hyperparameter tuning for deep learning
Real-world use: image recognition, NLP tasks

Text preprocessing: tokenization, stemming, lemmatization
Stopwords removal and text normalization
Bag-of-Words, TF-IDF, and word embeddings (Word2Vec, GloVe, FastText)
Named Entity Recognition (NER) and POS tagging
Sentiment analysis and text classification
Topic modeling (LDA, NMF)
Transformers and BERT architecture
Hugging Face pipelines and fine-tuning models
Real-world NLP projects: chatbot, review sentiment, keyword extraction

Time series components: trend, seasonality, noise
Stationarity, autocorrelation, and ACF/PACF analysis
ARIMA, SARIMA, SARIMAX models
Facebook Prophet for forecasting
Feature engineering for temporal data
Handling irregular time intervals and missing timestamps
Rolling windows and lag features
Deep learning for time series (LSTM, Temporal CNNs)
Real-world use: stock prediction, demand forecasting