Signed in as:

filler@godaddy.com

Signed in as:

filler@godaddy.com

This series consists of two courses: Introduction to Machine Learning and Introduction to Python. Learn from assignments/quizzes, our interactive discussion board, and weekly graduate student led lab sessions. By the end of this course series, you will have sufficient knowledge of programming algorithms and data set manipulation to be able to design your own simple and well-designed program.

Introduction to Machine Learning is dedicated to teaching the differences between machine learning algorithms and which one is more suitable given the problem we face. Across a few weeks, students will delve into programming and understand the mindset of the world's brightest minds. By combining the course material with practical examples, students will develop a strong foundation for our following course and for future pursuits. After concluding the course, students will be able to understand the evolution of machine learning algorithms, the pros and cons of each one and which one is more suitable to a problem.

Apply to our program below to get a head start in your programming pursuits!

This course aims to teach how to use python to write simple codes and programs, manipulate datasets, and to use Python for machine learning. Spreading across multiple modules, this course teaches not only the internal functions of Python, but also the extent of comprehending datasets from certain tools. Through each lecture, we will combine the knowledge learned with practical examples to hone the information to build the strong foundation. After taking this course, students will be able to write Python script (using functions, loops and statements) to solve problems, and be able to analyze datasets using variety of Python tools and libraries (NumPy and Pandas).

- Master the basics of information science and use the programming language Python for data analysis
- Attend weekly
**live webinars**to interact with instructors and other students - Receive a
**certificate of completion**with grades at the end of course - Receive a
**recommendation letter**upon the completion of Stage 2

Hi, I'm Odysseas Drosis. I graduated from the National Technical University of Athens with a degree in computer science. To continue my studies, I joined Cornell University for a Master's of Engineering in Computer and Information Sciences. Now, I am a Ph.D. student to study machine learning at EPFL in Switzerland. I worked for Nokia Corporation, a telecommunications company. I hope to dispense academic knowledge and provide insight into careers in information science for students.

My name is Yuqing Wu. I’m a graduate student majoring in Financial Engineering at Cornell University. I received my Bachelor of Science in Actuarial Science/Applied Statistics/Accounting from Purdue University. With my past working experiences in consulting and finance-related fields, I have accumulated solid knowledge in applying Python and Machine Learning in solving real-world challenges. I‘m now working on incorporating Machine Learning in data analysis and portfolio optimization, hoping to discover a new way of greeting the era of Machine Learning. I am excited to be working as an instructor for IvyOnline.

**Week One: Different Aspects of Machine Learning and Training Basic Linear Algorithms**

A general overview of topics covered during this course and what machine learning is. Feature extraction and ways to measure performance of the algorithms. Introduction to unsupervised learning. What is linear regression and the intuition behind this idea? In which cases can we use this technique?

**Week Two: More on Linear Algorithms, Metrics, and Non-linear Models **

Metrics to measure performance. What other algorithms belong to the field of linear algorithms? Why do we need some non-linear algorithms? Can we use the same metrics for these algorithms? What is the kernel trick and the intuition behind this great idea to connect linear and non-linear algorithms.

**Week Three: Introduction to Unsupervised Learning **

Cross validation technique and why we need this to improve the performance. What is boosting (high level idea) and how can we avoid overfitting?What is unsupervised learning?

**Week Four: Unsupervised Learning and Applications of the Course**

The main differences between supervised and unsupervised learning. Most common algorithms for this field (like clustering and dimensionality reduction), sum up what was covered during the course.

**Week One: Introduction to Python/Basic Python Elements/Python Lists, Sets and Dictionaries**

What is Python, what are Python advantages over other languages, what are some Python environments (including Jupiter Notebook and Google Colab)? What are some basic Python data types? What are Variables, Math operators, Logic operators, Input/output operators and comments? What are Python lists, sets and dictionaries? When to use each of them? What are their characteristics?

**Week Two: Python Loops and Statements**

What is If/else statements? What is for loop and while loops? What are the differences? When do we use each of them?

**Week Three: Python Functions**

What is Python functions? How to construct Python functions to simplify calculations?

**Week Four: Python Numpy, Pandas Libraries**

What are Python libraries? What are Numpy and Pandas? What are they used for? How to use them to analyze dataset?

**Bonus: Using Python to Read Files and Analyze Datasets **

How do we read file in Python? How do we analyze datasets using python?

In order to make this problem a binary classification problem, we will solve the following question: * Will the temperature increase or not the following day?* Students will understand how time-series problems work and how to best approach such problems to achieve better results. By learning how to construct and evaluate algorithms, students will learn the logical framework for experts in this field. Time series problems are commonly used in finance, especially during trading where we would predict future values from previous day data.

Students will analyze a real non-time series problem and understand how to solve binary classification problems. Given the features, they will be able to solve the binary problem with the techniques, such as boosting and cross validation, we learned during stage 1. Through these techniques, you will understand not only how to construct and edit algorithms, but also how to improve their performance. After working through this case study and the temperature problem, students will have a strong understanding of machine learning problems as these problems address fields of machine learning. .

This problem is about finance, but more specifically banks. Given some features, we are trying to predict if the client has **subscribed to a long-term deposit or not**. As usual this is going to be a binary classification problem and is quite useful for banks to achieve great accuracy on it. Students who are interested in finance are welcome to try to solve this problem.

This is a binary classification problem and students can use not only the algorithms we introduced in the first stage, but also the techniques to improve their performance.

With this case study being one of the more difficult problems of stage 2, students will be able to ask harder and more advanced questions about this topic. Ranging from the basic idea of how age affects death rate to which feature impacts the pandemic the most, students can delve deep into machine learning to see what other aspects should be considered. Students will develop a strong intuition and deep understanding of datasets and machine learning algorithms. Also, students will understand how to select the best metrics in order to evaluate the performance of the algorithm. By building their logical framework, students will be able to ask questions about a difficult topic and output their perception of the problem constructed by in depth analysis of the issue.