Popular repositories

1. Data-science-for-marketing Data-science-for-marketing Public

   In this code, the customer segmentation with respect to Recency, Frequency, Monetary value, is realized. This code is written with the aim to explore how data science can be applied in marketing, e…

   Jupyter Notebook 1

2. Topic-modelling-with-LDA-on-Russian-Texts Topic-modelling-with-LDA-on-Russian-Texts Public

   Let's find out 25 most common topics (their keywords) from 20000 texts using LDA topic modelling. We will test whether filtering of the corpus by tf-idf will improve accuracy of topic estimation. F…

   Jupyter Notebook 1

3. Machine-learning-classifications-and-predictions Machine-learning-classifications-and-predictions Public

   This code is aimed to explore abilities of Machine learning predictive techniques. It is based on Kaggle dataset https://www.kaggle.com/uciml/breast-cancer-wisconsin-data and focuses on minimizing …

   Jupyter Notebook

4. Deep-learning-on-Keras-framework Deep-learning-on-Keras-framework Public

   This code is written in the workframe of getting familiar with deep learning techniques on Keras, and is aimed to predict customer churn based on characteristic data about the customers.

   Python

5. Convolutional-Neural-Networks-Image-recognition Convolutional-Neural-Networks-Image-recognition Public

   This code is written in the workframe of studying one of the application of neural nets: image recognition. The dataset contains images of dogs and cats in separate folders, as well as the folder w…

   Python

6. Credit-card-fraud-prediction-overcoming-the-data-imbalance Credit-card-fraud-prediction-overcoming-the-data-imbalance Public

   In this repository, there are results of applying different techniques on a highly imbalanced data in order to improve prediction capabilities of the model.