|
HU Credits:
3
Degree/Cycle:
2nd degree (Master)
Responsible Department:
Business Administration
Semester:
1st Semester
Teaching Languages:
Hebrew
Campus:
Mt. Scopus
Course/Module Coordinator:
Lev Muchnik
Coordinator Office Hours:
By appointment
Teaching Staff:
Prof Lev Muchnik
Course/Module description:
This course will introduce the students to many concepts and techniques used in machine learning with the emphasis on application of the learned material to common problems faced by modern businesses. The course provides hands-on experience and trains students to pick the best tools for the specific problem in hand. Beyond application of the machine learning algorithms and interpretation of the results, the students will acquire skills necessary to collect, manage, clean and prepare the data for the analysis.
Course/Module aims:
Learning outcomes - On successful completion of this module, students should be able to:
On successful completion of this course, the students should be able to identify opportunities for application of machine learning techniques to real-world problems, define research questions and plan the study, organize and prepare the data for the analysis, select the most adequate technique, interpret and visualize the obtained results. Finally, the students will learn to establish data-driven decision-making procedure that will culminate in actionable steps aimed at improving company’s performance.
Attendance requirements(%):
80
Teaching arrangement and method of instruction:
The course will combine lectures demonstrating variety of machine-learning techniques with practical assignments that leverage these methods.
Course/Module Content:
Classical Classification Algorithms
• Support Vector Machine (SVM)
• Naïve Base classifier
• XGBoost
Intro to Deep Learning (Neural Networks):
• Perceptron
• Back propagation
• Fully Connected layer
• Types of networks:
• Feed forward
• Convolutional Neural Networks (CNN)
• Long-Short Term Memory (LSTM)
• Transfer learning
• Attention learning
• Classifying texts and images
• Packages: ImageNet, Keras
• Video: Object detection & tracking – YOLO
Natural Language Processing (NLP)
• TF-IDF
• Word Cloud
• Mutual Information
• Topic Modelling
• Word embedding (genism)
• Preprocessing text
• Stemming & Lemmatizing
• Cleaning (number, punctuation)
• BeautifulSoup
• MS Office Documents
• Text Classification
• Text Summarization
• Packages: SPACY/ NLTK/textract
• Recent Algorithms (BERT, XLNet)
Numerical Optimization - Heuristic Methods
• Gradient Descent
• Simulated Annealing
• Genetic Algorithm
• Ant Colony
Modelling and Numerical Simulations for hypothesis testing
• Agent-based modelling
Time series analysis & Prediction
• fbProphet
Anomaly (Outlier) Detection
Network Analysis
• NetworkX
Databases:
Structured Query Language (SQL)
Document Databases (MongoDB)
Crawling: Scrapy / Web APIs
Parsing: regex, xpath
Parallel Code: Multiprocessing
Logging
Large (binary) files
Git & code management
Additional topics
• Biases in data
• Causal Inference
• Unbalanced classes & missing data
• Distance Measures
• Confidence intervals, bootstrapping
• Grid Search
• Model Validation
• Ethics of Data Science
Required Reading:
The reading material will be provided in the class.
Additional Reading Material:
Course/Module evaluation:
End of year written/oral examination 0 %
Presentation 0 %
Participation in Tutorials 10 %
Project work 0 %
Assignments 50 %
Reports 0 %
Research project 40 %
Quizzes 0 %
Other 0 %
Additional information:
Room 5102B
School of Business Administration
The Hebrew University of Jerusalem
Mt. Scopus
|
