Machine Learning Workshop
This wokshop spans 32 hours over 4 weeks.
Two times per week ,(Saturdays&Tuesdays) each session 4 hours from 8 am t o 12 pm. Starting date: 5th August 2017 Location: iHub Studio, Ain Shams University CHEP Training Credit: 4 weeks Cost: Free but you will pay 200 EGP as a deposit (to be returned to you when you complete the workshop till the end). |
Instructor
Eng. Rana ElRobi
|
Registration Status
|
OVERVIEW
This workshop targets 3rd and 4th year students from computer science/computer Engineering,
It gives overview on main machine learning algorithms families and how to use each algorithm.
GOALS
This workshop targets 3rd and 4th year students from computer science/computer Engineering,
It gives overview on main machine learning algorithms families and how to use each algorithm.
GOALS
- Students know the main branches of Machine learning Algorithms Tree.
- Learn when and how to use each algorithm and its pros and cons.
- Practice on each algorithm even on a simple dataset to gain the experience of using the algorithms and see its results on the given data.
- Get familiar with using R scripts.
- Learn how to evaluate each algorithm result.
- Know where they can find data resources online.
- Learn how to make feature selection from the given data
COURSE CONTENT
Sat,5 th August
● Introduction about Machine Learning and Data science ● Review Main basics in Linear Algebra (needed in this course) ● Linear Classifiers ( Discriminant Function Classifiers ) |
[Practical]
➢ Getting started with R scripts ○ Load data and read it in vars ➢ Apply Least Square Classifier on given Data ➢ Apply Fisher Classifier on given Data |
Tues, 8 th August
● Review on main basics in Probability (needed in this Session) ● Linear Classifiers ( Probabilistic Models |
[Practical]
➢ Apply Naive Bayes Classifier on given Data ➢ Apply Logistic Regression |
Sat, 12 nd August
● Non Linear Classifiers ○ Knn ○ Weighted knn ○ Decision Tree |
[Practical]
➢ Apply Knn Classifier on given Data ➢ Apply weighted Knn Classifier on given Data ➢ Apply Decision Tree Classifier on given Data |
Tues, 15 th August
● Clustering Techniques ○ K-means ○ K-medoid ○ How to choose best K |
[Practical]
➢ Apply K-means on given Data ➢ Apply K-medoid on given Data |
Sat, 19 th August
● Clustering Techniques ○ Hierarchical Clustering ○ Specular Clustering |
[Practical]
➢ Apply Hierarchical Clustering technique on given Data ➢ Apply Specular Clustering technique on given Data |
Tues, 22 nd August
● Feature Selection ○ T-test Method ○ Forward and backward wrapper method |
[Practical]
➢ Apply ”T-test” method for feature selection on “ISOLTE” Dataset |
Sat, 26 th August
● Principal Component Analysis ○ Linear PCA ○ Nonlinear PCA |
[Practical]
➢ Apply ”Linear PCA” to do feature transformation ➢ Apply ”Nonlinear PCA” to do feature transformation |
Tues, 29 th August
● Support vector machine ○ Linear SVM ○ Nonlinear SVM |
[Practical]
➢ Apply ”Linear SVM” on given data. ➢ Apply ”Nonlinear SVM” on given data. |
Thurs,31 st August
● Ensemble Learning ○ Bagging ○ Boosting ○ Random Forest Technique |
[Practical]
➢ Apply ”Bagging” , “Boosting” and ”Random forest” on MNIST Dataset |
Sat, 9 th Sep
● Neural Network ○ Build small networks ○ How to build complex Network ○ Into to Deep Learning ○ Resources to Deep learning |
[Practical]
➢ Apply Neural Network on given Data. |
Tues, 12 nd Sep
● Text Mining Session (set by another Research assistant )
● Text Mining Session (set by another Research assistant )