Projects completed while studying at "Data Scientist Plus" program by Yandex.Practicum (2021-2023)
State-recognized Diploma of professional retraining (PDF): English / Russian
| Project | Description | Tools / Libraries | Status | |
|---|---|---|---|---|
| 01 | Big cities music preferences | Check several statistical hypotheses regarding music preferences of people living in Moscow & Saint-Petersburg | pandas |
Done |
| 02 | Analysis of bank's borrower reliability | Analyze if the marital status and the number of children of the bank's client affect the fact of loan repayment on time | pandas |
Done |
| 03 | Analysis of advertisements for the sale of apartments | Based on data from Yandex.Estate, determine the market value of real estate in Saint-Petersburg | pandas, matplotlib |
Done |
| 04 | Study of data on film distribution | Perform analysis to study the film distribution market and identify current trends | pandas, numpy, matplotlib, seaborn |
Done |
| 05 | Determination of a prospective tariff for a telecom company | Telecom company's commercial department needs analysis to understand which tariff brings in more money | pandas, numpy, scipy, matplotlib, seaborn |
Done |
| 06 | Mobile tariff recommendation for a client | Classification model is required to select the appropriate mobile tariff for clients | pandas, numpy, matplotlib, seaborn, sklearn |
Done |
| 07 | Bank's customer churn modelling | Based on historical data about clients' activities, it is necessary to predict whether some particular client will stop being bank's client in the nearest future or not | pandas, numpy, matplotlib, seaborn, sklearn |
Done |
| 08 | Choosing a location for an oil well | Having data about oil samples in three geographoc regions, build a model to choose the most profitable oil well location | pandas, numpy, matplotlib, seaborn, sklearn |
Done |
| 09 | Predicting rejection of hotel reservation | Develop a model that predicts hotel booking rejection and find out if the profit from such model would cover model's development expenses | pandas, numpy, matplotlib, seaborn, sklearn |
Done |
| 10 | SQL Basics | Train writing basic SQL queries | SQL |
Done |
| 11 | Git and Command-Line | Practice using Git and Linux command-line workflow | git, cmd |
Done |
| 12 | California housing cost prediction | Build a linear regression model on California housing data in 1990-s to predict median cost of a house in a residential area | pandas, numpy, pyspark |
Done |
| 13 | Linear algebra to protect personal data | Needed to protect the data of clients of an insurance company by developing a data transformation method that makes it difficult to recover personal information | pandas, numpy, sklearn |
Done |
| 14 | Cars cost determining | Based on historical data about technical characteristics, completeness and prices of cars, it's needed to build a model to determine car costs | pandas, numpy, matplotlib, seaborn, sklearn, catboost, lightgbm |
Done |
| 15 | Advanced SQL | Write 10 more advanced SQL queries from Jupyter environment | SQL |
Done |
| 16 | Star temperature prediction | Having characteristics of 240 cosmic stars already studied, create a neural network to determine the temperature on the surface of the discovered stars |
pandas, numpy, matplotlib, seaborn, sklearn, pytorch |
Done |
| 17 | Risk of road accident assessing | Create a system that could assess the risk of a road accident along the selected route; find out whether it's possible to predict an accident based on the historical data of one of the regions | SQL, pandas, numpy, matplotlib, seaborn, sklearn, sqlalchemy, lightgbm, catboost |
Done |
| 18 | Forecasting taxi orders | Build a model to predict the number of taxi orders for the next hour to attract more drivers during the peak period | pandas, numpy, matplotlib, seaborn, sklearn, statsmodels, lightgbm |
Done |
| 19 | Classification of comments whether they are toxic or not | Having a labeled English comments dataset with toxity markup, create a model to classify them into positive or negative | pandas, numpy, matplotlib, seaborn, nltk, spacy, lightgbm, afinn, nrclex |
Done |
| 20 | Determining age of buyers by their photos | Build a model that will determine the approximate age of a person from a photograph by using a labeled dataset of people photographs | pandas, matplotlib, seaborn, keras |
Done |
| 21 | Search images by text query | Build a model that is capable to get a textual description of some scene, and return several photos with the same or similar scene | pandas, numpy, matplotlib, seaborn, sklearn, pytorch, transformers, torchvision |
Done |
| 22 | Predict telecom contract termination | Build a model that will predict whether the subscriber will terminate the contract with telecom company or not | SQL, pandas, numpy, matplotlib, seaborn, sqlalchemy, sklearn, lightgbm, catboost, pytorch |
Done |