Start Date: 05/04/2022
Course Overview
Data scientists use a set of algorithms which enables computers to solve problems that are classified on a higher complexity level than traditional algorithms. Examples of such cases are: to predict a consumer behavior by its past choices, recognize a person within an image, “understand” written text, to predict a system failure or a cyber-attack.
Machine learning algorithms allow the computer to train and learn from its own mistakes and thus perfect its performance on new data.
This course gives the basis of understanding the data scientist environment, focusing mainly on common frameworks in order to enable selecting the appropriate approach to the problems at hands.
We will review various use cases and implement appropriate models and tools.
Who should attend?
Managers and architects who like to understand the different problems that are suitable for machine learning and exercise different frameworks.Prerequisite:
Basic programming skills in C, Java or any other language.
Course Outline:
Day 1:
1. Introduction to data science
• Examples and use cases
• Statistics 101
• Machine learning introduction
2. Data preparation using various tools
• Exploratory data analysis
• Cleaning the data
• Filtering and scaling
• Outliers and null values
• PCA
3. Running machine learning algorithms
• Regression and decision trees
• Statistical reasoning
• Clustering
• Weka Introduction
4. Mini project Part A: Recommendation System
• Data Preparation
• Feature selection
Day 2:
5. Machine learning in cloud environment, Big Data
• Classification
• Association Rules
• Decision Trees
6. Validation of Results
• Standard metrics
• ROC curve analysis
7. Mini Project Part B: Recommendation System
• Estimation of different models
• Demo
8. Summary