Feature Explainability & Traceability
Build features that are easily explained and that provide in-depth visualizations back to source data.
Automatic Transactional Aggregation
dotData automatically aggregates and combines data across transactional information.
Time-Series on Steroids
Automatically aggregate data across different time resolutions without human intervention.
Automatic Leakage Detection
Automatic detection of leaky features saves you time and frustration.
Large Data Dimensionality
dotData can process data that is both diverse as well as varied in nature to build richer, more insightful features.
Cleanse Your Data, Automatically
Automatically cleanse and prep your data for AI and ML algorithms before feature engineering even takes place.
Iterate Features In Record Time
Leverage automatic statistical data on each feature to analyze & evaluate features and iterate quickly and easily.
Support Broad Data Types
Explore and discover multimodal patterns by wrangling and combining relational, transactional, temporal, geo-locational, and text data, automatically.
Millions of Rows, Dozens of Tables
Discover and evaluate features across millions of rows, dozens of tables and thousands of columns - automatically.
Discover 100X more Features and Build Better Models
Rapidly iterate and test thousands of features for your ML & AI models with automatic data cleaning, hypothesis testing, and feature selection.
Trusted by global brands
Made by Data Scientists for Data Scientists
Request a product walkthrough
FAQ
Which cloud providers does dotData integrate with?
dotData works seamlessly on major cloud data platforms such as Databricks, Snowflake, AzureSynapse, Amazon Redshift, and Amazon EMR
Can I manually select the features to use?
Yes. dotData provides you with full access to sort and filter the features generated in order to identify the most relevant features for you modeling purpose.
Can dotData handle unstructured data such as XML, or JSON?
Within dotData you can set up pipelines to send unstructured or nested data directly into a Spark dataframe so that it is ready to use.
Does dotData perform transformations to discover features?
Yes. dotData performs a variety of temporal, categorical, geographic, textual, and mathematical transformations in order to explore and evaluate all possible feature hypothesis. You can even add your own!
Can dotData connect to our SQL database? Do we need to write a large amount of queries to generate the features?
Yes dotData can connect to your SQL database. You do not need to write any SQL queries in order to build your features or run the models.
Can I connect the features to my feature store?
dotData generates feature pipelines that can be used to automatically process raw source tables and push results into your feature stores
