Hello!
My name is Stanislav Tyulyagin, I am a Data Scientist and Python Developer. AI master's student at HSE '26.
In the Data_Sciense_Portfolio repository, you can explore my projects in the fields of DS and DA.
In the Python-developer list, you can find and examine my Python development projects in more detail.
In 2022, I completed the Data Scientist course at Yandex.Practicum, as well as the Deep Learning School course from MIPT. In July 2023, I finished the Python-backend course at Yandex.Practicum. I enjoy challenging tasks, am eager to learn new things, and can work both in a team and independently. I am self-organized, goal-oriented, meet deadlines, and can work remotely for extended periods without compromising productivity.
I enjoy analyzing and finding non-obvious patterns. I plan to continue developing and working in this field to make the world better through machine learning and Python development.
Currently, I am actively looking for a job and considering offers.
Contacts
- Telegram: https://t.me/tyulyagins
- E-mail: [email protected]
| Title | Description | Tools | Keywords |
|---|---|---|---|
| Comparison of music preferences | Comparison of music preferences between Moscow and St. Petersburg residents based on Yandex.Music data. | numpy pandas |
EDA, analysis |
| Bank loan reliability analysis | Data processing and analysis to identify factors influencing a bank client’s solvency. | pandas numpy matplotlib |
EDA, analysis |
| Real estate market analysis | Analyzing data and identifying factors affecting apartment prices. | numpy pandas matplotlib |
EDA, analysis, feature engineering |
| Telecom tariff profitability analysis | Determining the most profitable telecom tariff based on customer usage data. Hypotheses were tested. | numpy pandas matplotlib scipy |
EDA, analysis, hypothesis testing, ttest |
| Video game market analysis | Determining the factors influencing the successful sale of a game. | pandas matplotlib scipy |
EDA, statistics, hypothesis testing, ttest |
| Title | Description | Tools | Keywords |
|---|---|---|---|
| Mobile tariff recommendation | Creating a classification model to recommend the appropriate mobile operator tariff. | pandas matplotlib sklearn statsmodels |
Time Series, regression, classification |
| Bank customer churn prediction | Building a model to predict customer churn based on client behavior data. The goal is to reduce costs on client retention. | numpy pandas matplotlib sklearn StandardScaler |
EDA, analysis, feature engineering, ROC-AUC, upsampling, unbalanced classification |
| Oil rig profit prediction | Creating a model to predict the profit from oil rigs. Based on the forecast, determining the best drilling region. | numpy pandas matplotlib sklearn |
EDA, analysis, regression, ROC-AUC |
| Gold recovery prediction from ore | Creating a model to predict the gold recovery coefficient from gold-containing ore. | pandas matplotlib sklearn CatBoost Optuna GridSearch |
EDA, regression, gradient boosting |
| Protecting personal data of an insurance company's clients | Protecting personal data of insurance clients. Developing a method to transform data so that the original information is hard to recover while maintaining the quality of linear regression. | pandas numpy Random numpy |
EDA, regression |
| Car price prediction model | Creating a recommendation system for car pricing based on its description. | pandas matplotlib sklearn lightgbm |
EDA, regression, gradient boosting |
| Taxi order prediction | Predicting the number of taxi orders for the next hour to attract more drivers during peak times. | pandas matplotlib CatBoost Prophet |
Time Series, regression, gradient boosting |
| NLP. Sentiment analysis of text | Predicting positive and negative comments from online store users. | numpy sklearn spacy torch BERT |
NLP, TF-IDF, classification |
| CV. Customer age prediction | Creating a neural network model to predict a person's age based on a photo. | tensorflow keras ImageDataGenerator ResNet |
CV, neural network, classification |
| Telecom churn prediction | Building a model to predict whether a user will leave a telecom service. | Pipeline phik Catboost XGBoost sklearn |
EDA, classification |
