Summary of Interpretable Data-driven Anomaly Detection in Industrial Processes with Exiffi, by Davide Frizzo et al.
Interpretable Data-driven Anomaly Detection in Industrial Processes with ExIFFI
by Davide Frizzo, Francesco Borsatti, Alessio Arcudi, Antonio De Moliner, Roberto Oboe, Gian Antonio Susto
First submitted to arxiv on: 2 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper proposes a novel approach, ExIFFI, for providing interpretable outcomes in anomaly detection (AD) processes. The conventional AD methodologies typically classify observations as normal or anomalous without providing insights into the reasons behind these classifications. This limitation is addressed by presenting the first industrial application of ExIFFI, which focuses on producing fast and efficient explanations for the Extended Isolation Forest (EIF) Anomaly detection method. The proposed approach is tested on two publicly available industrial datasets, demonstrating superior effectiveness in explanations and computational efficiency compared to other state-of-the-art explainable AD models. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Anomaly detection is important in industries because it helps find problems early, which can prevent accidents or reduce waste. Right now, most anomaly detection methods just say whether something is normal or not, but they don’t tell us why. This makes it hard to understand what’s going on and make good decisions. The paper presents a new approach called ExIFFI that provides explanations for anomalies, so we can understand the reasons behind them. It uses a method called Extended Isolation Forest (EIF) and is tested on two real-world datasets. The results show that ExIFFI is better than other similar methods at providing explanations and using less computer power. |
Keywords
» Artificial intelligence » Anomaly detection
