eScience approach for applied machine learning

Description

Options

  • Course level: 400
  • Start date: as required
  • Duration time: 5 days

Price: 5000 zł

  • Open training
  • On-site (costumer's premises)
  • Customer’s computer

Class

Plan

The basic training is comprised of 12 modules. Just like all the other courses on our offer, it has been designed by our experts, therefore we can freely customize it to suit the individual needs of the participants. We would like to encourage you not only to choose modules which you find relevant for your needs, but also to feel free to contact us with your suggestions and questions you would like to hear answered during the course.

Modules

Duration: Level:
Module 1

Data Science in the Cloud
  • Introduction to data science and machine learning
  • Data Scientist
  • Toolbox: SQL, R and Azure ML

120 minutes 300
Module 2

Introduction to R
  • R and R Studio
  • Data types
  • Objects
  • Vectorized operations
  • Functions
  • Modules
  • Plotting systems

240 minutes 400
Module 3

Data Analysis and Statistical Inference
  • Descriptive statistics
  • Probability and distributions
  • Inference
  • Regressions

240 minutes 400
Module 4

Azure Machine Learning
  • Azure Machine Learning
  • Data analysis methodology
  • Supervised and unsupervised learning algorithms
  • Publishing Scoring Experiments
  • Prediction Queries

120 minutes 300
Module 5

classifiers (experiment)
  • Data assessment
  • Summary statistics
  • Bootstrapping, Bagging and Boosting
  • Decision Trees
  • Ensembles - combining classifiers (Random forest, Decision forest, Boosted decision trees)
  • Model evaluations (accuracy, precision, recall, F-Score, AUC, average log loss, and training log loss metrics)

180 minutes 400
Module 6

Regressors (experiment)
  • Linear regression
  • Data Visualization
  • Data Tidying
  • Regression models (Logistic regression, Neural networks)
  • Model Evaluation (mean absolute error, root mean squared error, relative absolute error, and relative squared error metrics)

180 minutes 400
Module 7

Reccomender (experiment)
  • Data Preparation
  • Recommender split
  • Bayesian recommender
  • Matchbox Recommender
  • Score and Evaluate Recommender
  • Azure ML Web Service

180 minutes 400
Module 8

anomaly detection (experiment)
  • Unsupervised Learning
  • Data Assesment and preparation
  • Clustering models
  • Clustering model evaluation
  • Published Scoring Experiment

180 minutes 400
Module 9

Fraud detection (experiment)
  • Data Evaluation
  • Data Enhancement
  • Advanced classifiers (Bayes points machines)
  • Model evaluation
  • Model Publication

180 minutes 400
Module 10

Text Retrieval and Classification (experiment)
  • natural language content analysis
  • text retrival and preprocessing
  • Extract and select features
  • Vector space model
  • n-grams
  • Support Vector Machine
  • Published scoring experiment

180 minutes 400
Module 11

Azure ML Studio components
  • Data retrival
  • Handling missing values
  • Oversampling
  • Cross validation
  • Model retraining
  • Clients

120 minutes 300
Module 12

ML Algorithms
  • Patterns recogition
  • Clustering methods
  • Nerual netoworks in deep
  • Graphs

120 minutes 300

Form

Ask Question

Register