# Mill.jl and JsonGrinder.jl: automated differentiable feature extraction for learning from raw JSON data

arXiv: https://arxiv.org/abs/2105.09107 · HTML: https://mlsanity.com/paper/2105.09107/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 5 of its 16 references that are in the map.
- First posted on arXiv 19 May 2021.

## Builds on (most cited first, 5 of 5)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Deep Sets](https://mlsanity.com/paper/1703.06114/) (2017)
- [EMBER: An Open Dataset for Training Static PE Malware Machine Learning Models](https://mlsanity.com/paper/1804.04637/) (2018)
- [Approximation capability of neural networks on spaces of probability measures and tree-structured domains](https://mlsanity.com/paper/1906.00764/) (2019)
- [Using Neural Network Formalism to Solve Multiple-Instance Problems](https://mlsanity.com/paper/1609.07257/) (2016)

## Built on by (most cited first, 2 of 2)

- [Nebula: Self-Attention for Dynamic Malware Analysis](https://mlsanity.com/paper/2310.10664/) (2023)
- [Avast-CTU Public CAPE Dataset](https://mlsanity.com/paper/2209.03188/) (2022)

## Abstract (arXiv)

Learning from raw data input, thus limiting the need for manual feature engineering, is one of the key components of many successful applications of machine learning methods. While machine learning problems are often formulated on data that naturally translate into a vector representation suitable for classifiers, there are data sources, for example in cybersecurity, that are naturally represented in diverse files with a unifying hierarchical structure, such as XML, JSON, and Protocol Buffers. Converting this data to vector (tensor) representation is generally done by manual feature engineering, which is laborious, lossy, and prone to human bias about the importance of particular features. Mill and JsonGrinder is a tandem of libraries, which fully automates the conversion. Starting with an arbitrary set of JSON samples, they create a differentiable machine learning model capable of infer from further JSON samples in their raw form.