| 25 |
Final Project |
Practicum DS |
In this project, we dig deeper into refining the feature set and move from a two- to a three-class model based on the results of running UMAP to better grasp the dataset |
numpy, pandas, matplotlib, seaborn, plotly, re, datetime, sklearn, scipy, keras, umap |
| 24 |
Intro to CV |
Practicum DS |
In this project, we demonstrate a simple model evaluating the age from a picture, while running it on an external GPU platform |
pandas, matplotlib, keras, inspect |
| 23 |
ML for Texts |
Practicum DS |
In this project, we apply different vectorization techniques to build a classification model for detecting negative reviews. |
numpy, pandas, matplotlib, math, re, sklearn, nltk, spacy, lightgbm, torch, transformers, tqdm |
| 22 |
Time Series |
Practicum DS |
In this project, we build a predictive model for a time series, while studying and tackling the issues of heteroscedasticity and autocorrelation. |
numpy, pandas, seaborn, plotly, time, sklearn, scipy, statsmodels, arch |
| 21 |
Numerical Methods |
Practicum DS |
In this project, we study the balance between quality, speed and time required to train a model of predicting the car price. |
numpy, pandas, seaborn, plotly, time, sklearn, scipy, lightgbm, catboost, xgboost |
| 20 |
Linear Algebra |
Practicum DS |
In this project we establish a customer’s personal data masking procedure to protect it during the analysis, with examples of finding similar profiles, classification, or regression analysis. |
numpy, pandas, seaborn, typing, sklearn |
| 19 |
ML in Business 2 |
Practicum DS |
This project demonstartes building a model based on a custom model quality metric. |
numpy, pandas, matplotlib.pyplot, plotly.express, seaborn, re, sklearn |
| 18 |
ML in Business |
Practicum DS |
This project demonstartes a simple model of oil well reserves prediction and applying the bootstraping technique to assess the risks and confidence level of the profit estimate. |
numpy, pandas, matplotlib.pyplot, plotly.express, seaborn, sklearn |
| 17 |
Supervised Learning |
Practicum DS |
In this project we need to predict whether a customer will leave a bank soon based on clients’ past behavior and termination of contracts. |
numpy, pandas, matplotlib.pyplot, plotly.express, seaborn, sklearn, optuna |
| 16 |
Intro to ML |
Practicum DS |
In this project we develop a model that would analyze subscribers’ behavior and recommend one of a mobile carrier new plans. |
numpy, pandas, matplotlib.pyplot, seaborn, sklearn.preprocessing, sklearn.model_selection, sklearn.linear_model, sklearn.ensemble, sklearn.metrics |
| 15 |
Final Project |
Practicum DA |
I took a task of investigating how much effective is the loyalty program in expanding the total customer value of existing customers and find ways to boost its effectiveness. |
numpy, pandas, matplotlib.pyplot, seaborn, plotly, scipy.stats, math |
| 14 |
Final Project - SQL |
Practicum DA |
I was approached by an aspiring book reading app and requested to analyse book reading preferences and to make recommendations for their future value proposition. |
sqlalchemy, pandas |
| 13 |
Final Project - A/B testing |
Practicum DA |
I received an analytical task from an international online store: analyze the results of an A/B test and evaluate whether it was carried correctly. |
numpy, pandas, matplotlib.pyplot, seaborn, plotly, scipy.stats, math |
| 12 |
Forecasts and Predictions |
Practicum DA |
I was a data analyst at a gym chain called Model Fitness. We are developing a customer retention strategy. To fight churn, Model Fitness digitized a number of its customer profiles. My task was to analyze them and come up with a customer retention strategy. |
numpy, pandas, matplotlib.pyplot, scipy.stats, seaborn, sklearn.preprocessing, sklearn.model_selection, sklearn.linear_model, sklearn.ensemble, sklearn.metrics, sklearn.cluster |
| 11 |
Automating Reporting with Tableau |
Practicum DA |
Every week, my colleagues needed answers to the same questions on trending videos, so we decided to automate the process of fetching the analytics they had to prepare. |
Tableau |
| 10 |
Integrated project 2 |
Practicum DA |
I worked at a startup that sells food products and my task was to investigate user behavior for the company’s app. First, we study the sales funnel and find out how users reach the purchase stage. Then, we look at the results of an A/A/B test to find out which set of fonts produces better results. |
numpy, pandas, matplotlib.pyplot, scipy.stats, seaborn, math, cufflinks, plotly |
| 9 |
Telling a Story with Data |
Practicum DA |
Me and my partners decided to open a small robot-run cafe in Los Angeles, our investors are interested to know whether the market conditions will allow scaling our success when the novelty of robot waiters wears off, so we prepared a market research. |
numpy, pandas, matplotlib.pyplot, scipy.stats, seaborn, math, cufflinks, plotly |
| 8 |
Making Business Decisions |
Practicum DA |
Together with the marketing department, we compiled a list of hypotheses that may help to boost revenue. The task is to prioritize these hypotheses and then to analyze the results. |
numpy, pandas, matplotlib.pyplot, scipy.stats, seaborn, math |
| 7 |
Business Analytics |
Practicum DA |
The task at hand is to make recommendations and help optimize marketing expenses for an app |
numpy, pandas, matplotlib.pyplot, scipy.stats, seaborn |
| 6 |
Data Collection and Storage |
Practicum DA |
The ultimate goal stated for the project was, based on the weather data from an open source and competitors’ data on taxi rides, to determine the influence of weather conditions on the demand for taxi rides |
requests, BeautifulSoup, json, pandas, matplotlib.pyplot, scipy.stats, seaborn |
| 5 |
Integrated project |
Practicum DA |
The goal stated for the project was to identify patterns that determine whether a game is going to succeed or not. This would allow the company to spot potential winners and plan an advertising campaign |
pandas, matplotlib.pyplot, scipy.stats, seaborn, sklearn.neighbors |
| 4 |
SDA 2 |
Practicum DA |
A telecom operator offered its clients two prepaid plans; the goal of the project was to find which one of the plans brings in more revenue, in order the commercial department could adjust the advertising budget |
pandas, matplotlib.pyplot, scipy.stats, seaborn |
| 3 |
SDA 1 |
Practicum DA |
The ultimate goal stated for the project is to determine which factors influence the price of a vehicle; i.e. to reveal sound assumptions with regard to cars’ characteristics which have the most impact the price of a particular vehicle |
pandas, matplotlib.pyplot, scipy.stats, seaborn |
| 2 |
EDA |
Practicum DA |
Explore data from a navigating app and find gas station chains which seem to have the longest refueling times |
pandas, matplotlib.pyplot, scipy.stats, seaborn |
| 1 |
Data Preprocessing |
Practicum DA |
Preprocess data from a bank’s loan division and test borrowers’ risk of defaulting to reveal sound assumptions with regard to customers’ characteristics which have the most impact on the probability of a particular customer to payout the loan |
pandas, matplotlib.pyplot, scipy.stats |