Building your first predictive model has never been easier.
Decanter is designed so that anyone can use it to build a predictive model in five minutes, whether or not you already have prior experience in data science. Simply upload your raw data, define a problem, and let Decanter do the hard work for you.
- Simple, easy-to-use interface
- Support for most data sources
- Efficient, automatic workflow
Automated data cleaning
Decanter provides a comprehensive list of methods for data cleaning and preparation such as missing value imputation, outlier handling, removing duplicates, and etc. What traditionally would take a data scientist numerous hours of manual work now only takes a few clicks on Decanter Cloud.
Automated feature engineering
Feature engineering is crucial in making machine algorithms work, but it is very difficult and time-consuming. It requires domain understanding, repeated trial-and-error, and strong analytics acumen to identify and engineer the important features. Decanter Cloud simplifies this process by providing a simple interface to utilize your domain knowledge and efficiently engineer features.
Automated model building
Decanter Cloud automatically builds predictive models by selecting and combining optimal machine learning algorithm(s) using a standard procedure that requires no human interference. Currently, Decanter Cloud supports over 100 machine learning algorithms for classification, regression, clustering, time series, and deep learning applications.