Course Description |
This course will teach the students how to start from scratch in answering questions about the real world using data. The model building process involves setting up ways of collecting data, understanding and paying attention to what is important in the data to answer relevant business questions, finding a statistical, mathematical or a simulation model to gain understanding and make
predictions. This process involves asking questions, gathering and manipulating data, building models, and ultimately testing and evaluating them.
R is the primary programming tool for reading datasets and model building and validation. |
Course Description in Turkish |
Bu ders öğrencilere verileri kullanarak gerçek dünyayla ilgili soruları yanıtlamaya sıfırdan nasıl başlayabileceklerini öğretecektir. Model oluşturma süreci, veri toplamanın yollarını oluşturmayı, ilgili iş sorularını yanıtlamak için verilerde neyin önemli olduğunu anlamayı ve bunlara dikkat etmeyi, anlayış kazanmak ve yapmak için istatistiksel, matematiksel veya bir simülasyon modeli bulmayı içerir.
|
|
Program Outcomes and Competences |
Level |
Assessed by |
1) |
Building on the skills acquired during the undergraduate degree, an improved and deepened level of expertise in the field of big data analytics related to machine learning. |
N |
|
2) |
Applied in-depth theoretical and practical knowledge in the fields of statistics, computation and computer science related to machine learning. |
N |
|
3) |
Extensive knowledge about the analysis and modeling methods used in machine learning and their limitations. |
N |
|
4) |
Ability to design and perform exploratory research based on analytics, modeling and experimentation; to generate solutions to complex situations encountered in this process and to interpret the results. |
N |
|
5) |
Ability to describe the analytics process and its results both verbally and in writing on national and international platforms within or outside of the field of machine learning. |
N |
|
6) |
Awareness of social, scientific and ethical values regarding the machine learning, processing, usage, interpretation and dissemination stages and in all related professional activities. |
N |
|
7) |
Professional awareness new and emerging applications in the machine learning field and an ability to demonstrate their uses. |
N |
|
8) |
Competence to act as a leader in multi-disciplinary teams, to develop big data-driven solutions to complex situations; to take responsibility. |
N |
|
9) |
Ability to communicate in English both verbally and in writing at European Language Portfolio General Level B2. |
N |
|
10) |
Understanding of social and environmental aspects of machine learning applications. |
N |
|
Week |
Subject |
1) |
Basic concepts (a simple example with basic eda and model validation) |
2) |
Basic concepts (a simple example with basic eda and model validation) |
3) |
Building a Linear Regression model with Python (a more detailed example with eda, feature engineering and cross-validation) |
4) |
Building a Linear Regression model with Python (a more detailed example with eda, feature engineering and cross-validation) |
5) |
Building a Tree Based Regression model with Python (an example with model tuning, model evaluation, cross validating time series data) |
6) |
Building a Tree Based Regression model with Python (an example with model tuning, model evaluation, cross validating time series data) |
7) |
Building a Boosted Tree Classifier model with python(an example with roc_auc, metrics, confusion matrix) |
8) |
Building a Boosted Tree Classifier model with python(an example with roc_auc, metrics, confusion matrix) |
9) |
Building an MLP model with Python (a basic implementation example with neural networks) |
10) |
Building an MLP model with Python (a basic implementation example with neural networks) |
11) |
Model Evaluation, Interpretation and Explainability (an example with feature importances, shap, comparison of different model outputs)Model |
12) |
Model Evaluation, Interpretation and Explainability (an example with feature importances, shap, comparison of different model outputs) |
13) |
Building an ensemble of models (diversity check, ensembles, stacks) |
14) |
Building an ensemble of models (diversity check, ensembles, stacks) |
15) |
Final Examination Period |
16) |
Final Examination Period |