Deep feature synthesis: Towards automating data science endeavors

By James Max Kanter, Kalyan Veeramachaneni. 2015

486 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Neural Architecture Search & AutoML

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Neural Architecture Search & AutoML

Citations in the corpus

Automated Machine Learning: State-of-The-Art and Open Challenges (2019, background): “For instance, Feature Tools [58] is a library for automated feature engineering which follows a deep feature synthesis algorithm that can work with relational databases making use of the entity, forward, and backward relations between tables in generating new higher level features iteratively that c”
Benchmark and Survey of Automated Machine Learning Frameworks (2019, methodology): “Some frameworks specialize on feature generation in relational databases (Kanter and Veeramachaneni, 2015; Lam et al., 2017).”
Survey on Automated Machine Learning (2019, methodology): “The only viable approach for candidate feature generation is deep feature synthesis (Kanter and Veeramachaneni, 2015).”
Automating Biomedical Data Science Through Tree-Based Pipeline Optimization (2016, background): “One recent example of automated feature construction is “The Data Science Machine,” which automatically constructs features from relational data sets via deep feature synthesis [9].”
Generalizing from a Few Examples (2019, methodology): “Another direction is to extend the AutoML methods of automated feature engineering [66], model selection [71] and neural architecture search [166] to FSL.”
One button machine for automating feature engineering in relational databases (2017, unknown): “However, recent work shows that, for a speciﬁc type of problem and data such as provided in relational databases, automation of feature engineering is achievable [5].”
Taking Human out of Learning Applications: A Survey on Automated Machine Learning (2018, methodology): “Existing works on this topic include Data Science Machine (DSM) [10], ExploreKit [9] and FeatureHub [22].”
Learning Feature Engineering for Classification (2017, background): “The Deep Feature Synthesis component of Data Science Machine (DSM) [Kanter and Veeramachaneni, 2015] relies on exhaustively enumerating all possible new features, then performing feature selection over the augmented dataset.”
ExploreKit: Automatic Feature Generation and Selection (2016, unknown): “The Data Science Machine [7] has the similar goal of generating new features to improve classification performance.”
Feature Engineering for Predictive Modeling using Reinforcement Learning (2017, methodology): “The Deep Feature Synthesis component of Data Science Machine (DSM) (Kanter and Veeramachaneni 2015) relies on applying all transformations on all features at once (but no combinations of transformations), and then performing feature selection and model hyper-parameter optimization over the combined ”
Human-AI Collaboration in Data Science (2019, methodology): “nown as ETL or Extract/Transform/Load) [42]. Feature engineering is often considered to be the most time consuming step of the data science process [33], and techniques such as Deep Feature Synthesis [35] and One Button Machine [43] automate the task of generating ML-ready features from multi-relati”
Machine learning pipeline for battery state-of-health estimation (2021, background): “Feature engineering is split into automatic feature generation or extraction through techniques such as neural network auto-encoders [40], [41], and manual feature construction based on domain knowledge [42], [43].”
FeatureHub: Towards Collaborative Data Science (2017, methodology): “Automated feature engineering: In [7], the authors develop an algorithm for automatically extracting features from relational datasets.”
Techniques for Automated Machine Learning (2019, methodology): “Automated Feature Engineering (AutoFE) FeatureRL [18] GP for Feature Engineering [30] Data Science Machine [14] ExploreKit [15] Cognito [19] AutoLearn [16]”
Evaluation of a Tree-based Pipeline Optimization Tool for Automating Data Science (2016, unknown): “One recent example of automated feature construction is the “Data Science Machine,” which automatically constructs features from relational databases via deep feature synthesis [11].”