Automatic and Adaptive Preprocessing for the Development of Predictive Models

Authors: Salvador, M.M.

Conference: 5th Annual DEC PGR Poster Conference


Data preprocessing is a high time consuming task in data mining because a lot of manual work and expert knowledge is needed to clean and prepare data for the building and exploitation of predictive models. My project will investigate how to deal with common problems in raw data such as missing values, outliers, noise or concept drift in online predictive systems and how to make it autonomously without human interaction. For that, an automatic and adaptive framework for the preprocessing of raw data is proposed.

Source: Manual

Preferred by: Manuel Salvador