Repository containing a portfolio of data science projects completed during the training courses at Yandex.Practicum. Presented in the form of Jupyter Notebooks and readme markdown files.
| Project | Description | Used libraries |
| Customer Churn Prediction for a Bank | The bank is experiencing a steady but noticeable decline in customer retention. To address this issue, the marketing team has determined that it's more cost-effective to retain existing customers than to acquire new ones. Hence, the aim of this project is to build a customer churn prediction model. The model will utilize historical data on customer behavior and contract terminations to forecast whether a customer is likely to leave the bank in the near future. | Python, Numpy, Seaborn, Matplotlib, Scikit-learn, Pandas, Jupyter Notebook, LightGBM, Exploratory Data Analysis (EDA), Preprocessing Data, Machine learning |
| Customer Churn Prediction for a Telco | The telecom operator aims to predict customer churn and provide incentives to customers intending to leave. They have gathered personal data, tariff information, and contract details. The objective of this project is to develop a churn prediction model using binary classification and supervised learning techniques with the available data. | Python, Numpy, Seaborn, Matplotlib, Scikit-learn, Pandas, Jupyter Notebook, LightGBM, Exploratory Data Analysis (EDA), Preprocessing Data, Machine learning |
| Semantic analysis of comments | To improve comment moderation efficiency, the online store automated toxicity assessment. The model was trained to classify comments as toxic or non-toxic, speeding up the moderation process. | Python, NLTK, Numpy, Scikit-learn, Pandas, Jupyter Notebook, LightGBM, NLP |