# Automated CVE Analysis: Harnessing Machine Learning In Designing Question-Answering Models For Cybersecurity Information Extraction

arXiv: https://arxiv.org/abs/2412.16484 · HTML: https://mlsanity.com/paper/2412.16484/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 28 references that are in the map.
- First posted on arXiv 21 Dec 2024.

## Builds on (most cited first, 8 of 8)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://mlsanity.com/paper/1910.01108/) (2019)
- [SQuAD: 100,000+ Questions for Machine Comprehension of Text](https://mlsanity.com/paper/1606.05250/) (2016)
- [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://mlsanity.com/paper/1906.08237/) (2019)
- [SecureBERT: A Domain-Specific Language Model for Cybersecurity](https://mlsanity.com/paper/2204.02685/) (2022)

## Abstract (arXiv)

The vast majority of cybersecurity information is unstructured text, including critical data within databases such as CVE, NVD, CWE, CAPEC, and the MITRE ATT&CK Framework. These databases are invaluable for analyzing attack patterns and understanding attacker behaviors. Creating a knowledge graph by integrating this information could unlock significant insights. However, processing this large amount of data requires advanced deep-learning techniques. A crucial step towards building such a knowledge graph is developing a robust mechanism for automating the extraction of answers to specific questions from the unstructured text. Question Answering (QA) systems play a pivotal role in this process by pinpointing and extracting precise information, facilitating the mapping of relationships between various data points. In the cybersecurity context, QA systems encounter unique challenges due to the need to interpret and answer questions based on a wide array of domain-specific information. To tackle these challenges, it is necessary to develop a cybersecurity-specific dataset and train a machine learning model on it, aimed at enhancing the understanding and retrieval of domain-specific information. This paper presents a novel dataset and describes a machine learning model trained on this dataset for the QA task. It also discusses the model's performance and key findings in a manner that maintains a balance between formality and accessibility.