TechnoVerse: Data and Artificial Intelligence Training with Python

Objective of the Training:

Designed for anyone looking to adapt to the revolutionary changes of the digital age and build a strong career in the fields of data science and artificial intelligence, this training aims to equip participants with the skills to perform powerful data analysis, implement advanced modeling techniques, and develop AI solutions using Python. Through an in-depth exploration of data analysis and machine learning algorithms in Python, participants will gain the expertise to make accurate and robust predictions on real-world datasets. This program offers not only theoretical knowledge but also hands-on skills, enabling participants to develop projects and stand out in the industry.

Training Scope:

1. Foundations of Data Science:

• Introduction to Python: Begin your journey into data science with a strong foundation in Python. Learn about variables, data types, functions, and essential libraries to build your programming skills.
• Data Manipulation with Numpy and Pandas: Develop the ability to understand and process data by organizing datasets with Pandas and performing numerical analyses using Numpy.
• Data Visualization with Matplotlib and Seaborn: Learn how to effectively visualize data and present analytical results in a clear and insightful visual format.

2. Machine Learning and Regression:

• Linear and Polynomial Regression: Discover how to model relationships within datasets using core regression techniques, exploring both linear and nonlinear trends.
• Ridge and Lasso Regression: Master regularization techniques to avoid overfitting and underfitting, ensuring more accurate and generalizable predictions.
• Random Forest and Gradient Boosting: Focus on powerful ensemble methods to build robust models and understand the differences between bagging and boosting algorithms.

3. Classification Models:

• Bagging and Boosting: Dive deep into ensemble classifiers such as Random Forest, XGBoost, and AdaBoost to develop highly accurate and reliable prediction models.
• Model Tuning and Hyperparameter Optimization: Learn how to improve your models by selecting optimal hyperparameters using techniques like GridSearchCV and RandomizedSearchCV.

4. Advanced Model Evaluation and Improvement:

• Model Performance Evaluation: Gain the skills to assess model effectiveness using metrics such as Confusion Matrix, ROC Curve, Precision-Recall Curve, and AUC.
• Cross-Validation and Model Enhancement: Apply cross-validation strategies to test generalization ability and implement improvements for achieving reliable and sustainable results.

5. Real-World Applications and Project Work:

• Hands-On Projects with Real Datasets: Strengthen your machine learning, data analysis, and modeling skills through applied projects. Gain practical experience by solving real business problems using real-world datasets.

Benefits of the Training:

• Elevate Your Career: This program provides a strong foundation in data science and artificial intelligence, equipping you with the latest technologies and preparing you for a future-oriented career.
• Acquire High-Demand Skills: Gain expertise in data analysis, machine learning, modeling, and hyperparameter optimization—skills that are highly sought after in today’s job market—boosting your competitiveness and employability.
• Make Better Decisions: Learn to extract meaningful insights from datasets and make informed decisions through visual analytics. These skills will benefit not only your professional life but also your day-to-day decision-making.
• Master AI and Data Science Tools: Get hands-on experience with powerful Python libraries such as Scikit-learn, XGBoost, Pandas, Numpy, Seaborn, and Matplotlib, and develop the capability to utilize the most essential tools in modern data science.

TechnoVerse: Data and AI Training with Python is a comprehensive program designed to prepare you for the technologies of the future and equip you with the most in-demand skills in the industry. Join us to gain the expertise that sets you apart in the world of data science and artificial intelligence—and take your career to the next level!

Day 1: Python Fundamentals and Data Analysis

Introduction to Python and Environment Setup

• Installing Python and working with IDEs (Jupyter Notebook, VSCode)
• Data types, variables, and basic Python structures

Numerical Computing with Numpy

• Numpy arrays and basic operations
• Mathematical functions and array manipulations

Data Manipulation with Pandas

• Understanding DataFrame and Series structures
• Reading and writing data (CSV, Excel formats)
• Data cleaning and preprocessing (handling missing values, data transformation)
• Advanced Pandas operations (groupby, pivot tables, merging datasets)

Day 2: Data Visualization – Descriptive Statistics with Matplotlib and Seaborn

Basic Visualizations with Matplotlib

• Line plots, bar charts, histograms, and scatter plots
• Adding titles, labels, and customizing colors and styles
• Including error bars to visualize variability in data points

Statistical Visualizations with Seaborn

• Boxplot: Visualizing data spread and identifying outliers
• Violin plot: Showing data distribution and density
• Pairplot: Exploring relationships among multiple variables
• Heatmap: Visualizing correlation matrices
• Scatterplot: Examining the relationship between two variables

Descriptive Statistics and Visual Representation

• Mean, Median, Mode: Visualizing measures of central tendency
• Variance and Standard Deviation: Using histograms and boxplots to understand distribution
• Quartiles (Q1, Q3) and IQR: Exploring data spread and central concentration visually
• Outliers: Detecting and visualizing extreme values using boxplots and scatterplots

 Day 3: Data Analysis and Advanced Visualizations

 Advanced Visualization Techniques:

• Pairplot: Visualizing relationships among multiple variables using Seaborn
• FacetGrid: Using FacetGrid to visualize data analysis across multiple categories
• Correlation Heatmap: Visualizing correlations between numerical variables using Pandas and Seaborn

Advanced Statistical Analyses and Visualization:

• Correlation and Regression: Adding scatterplots and regression lines to understand data correlation
• Histograms and Distribution Plots: Analyzing data distribution and density
• Outlier Analysis with Z-score and IQR:
• Visualizing outliers using boxplots and scatterplots
• Identifying and visualizing outliers with Z-score and IQR (interquartile range)

Day 4: Data Preprocessing and Feature Selection

Data Cleaning:

• Handling Missing Data: Detecting missing values and filling them appropriately (mean, median, interpolation, etc.)
• Data Formatting: Converting data types (dates, text, categorical variables), identifying and correcting incorrect data
• Methods for Handling Missing Values:
• Filling missing data with mean, median, or forward/backward fill
• Dropping rows or columns with missing values

 Data Transformation:

• Data Normalization and Scaling: Making data more homogeneous
• Using StandardScaler (standardization) and MinMaxScaler (normalization)

 Encoding Categorical Variables:

• One-Hot Encoding: Converting categorical data into numerical format
• Label Encoding: Converting categorical variables into ordered numerical values

Feature Selection and Advanced Feature Transformation:

• Feature Engineering
• Feature Selection Methods

Day 5: Fundamentals of Machine Learning and Regression Models

• Introduction to Machine Learning:

• Differences between supervised and unsupervised learning
• Creating training, test, and validation sets

• Regression Analysis:
• Linear Regression: Analyzing relationships with simple linear regression, training the model, and making predictions
• Polynomial Regression: Modeling nonlinear relationships in the data
• Model Evaluation: Calculating performance metrics such as RMSE, MSE, MAE, and R-squared

Days 6 and 7: Core Methods in Classification and Regression Models

Hyperparameter Optimization:

• GridSearchCV and RandomizedSearchCV
• Testing model generalization capability using Cross-Validation
• K-fold Cross-Validation
• Stratified K-fold

Overfitting and Underfitting Analysis

Model Evaluation and Improvement

Model Evaluation Metrics

• For Classification Models:
  • • Confusion Matrix: A table that shows model performance with correct and incorrect classifications
    • Accuracy: The proportion of correct predictions
    • Precision, Recall, and F1-Score: Metrics that evaluate precision, sensitivity, and the balance of the model
    • ROC Curve and AUC (Area Under Curve): Visual evaluation of classification success, especially for imbalanced datasets
    • AUC: The area under the ROC curve, used to measure the model's accuracy
    • Precision-Recall Curve: Evaluation of model performance for imbalanced datasets; especially important in cases of class imbalance
• For Regression Models:
• • • Mean Absolute Error (MAE)
    • Mean Squared Error (MSE) and Root Mean Squared Error (RMSE)
    • R-squared (R²)
    • Adjusted R-squared 

Day 8: Classification Models – Core Methods

Bagging Algorithms

• Random Forest
• Bagging Classifier

 Boosting Algorithms

• AdaBoost
• Gradient Boosting
• XGBoost (Extreme Gradient Boosting)
• HistGradientBoosting
• Stochastic Gradient Boosting (SGD Boosting)
• LightGBM (Light Gradient Boosting Machine)

 Days 9 and 10: Classification and Regression Models – Advanced Algorithms

• Clustering Algorithms
• Support Vector Machines (SVM)
• Classification Algorithm Applications
• Linear Regression
• Polynomial Regression
• Ridge and Lasso Regression
• Decision Tree Regression
• Random Forest Regression
• Gradient Boosting Regression
• Support Vector Regression (SVR)Support Vector Regression (SVR)

Eğitim Takvimi:

Weekdays 10.00-17.00

Eğitmen:

Doç.Dr. Öyküm Esra YİĞİT

Doç.Dr. Ümit GÜNEŞ

Dr.Öğr. Üyesi Kaan ÜNLÜGENÇOĞLU