Main work lines

Bellow are listed some data science tasks we carry out frequently. However, we love challenges, thus if your project requires less frequent or experimental techniques, we will be pleased to hear your proposal.

Churn Analysis

Identify higher value customers, determine what actions typically precede a customer or sale loss, or understand what factors influence customer spending. When you improve customer retention, you substantially improve the bottom line.

Recommender Systems

System able to estimate the rating or preferece a user would give a certain item. Nowadays recommender systems are widely used for movies, music, news, books, etc. Although, recommender systems can also be useful for experts, restaurants, financial services, life insurances, persons (online dating), etc.

Text mining and Sentiment Analysis

By means of natural language processing, text analysis and computational linguistics it is possible to identify and extract subjective information from raw source materials (as books, news, or even voice recordings). Sentiment analysis is widely applied to product reviews and social media in many and diverse situations, ranging from marketing to customer service.

Risk Assessment

Given a certain action or event, this task consist in estimating the magnitude of event's consequences as well as its probability to occur.

Fraud Detection

Fraud is a billion-dollar business and its interest increases year by year. The PwC global economic crime survey of 2009 suggests that close to 30 percent of companies worldwide have reported being victims of fraud in the past year.

Survival Analysis

Survival analysis study the expected amount of time until one or more events happen, such as death in biological organisms or failure in mechanical systems.

Main research lines

Automatic machine learning

Spatial Tree based methods

Meta-learning

General Purpose Computing on GPU

Non-parametric clustering