---
license: cc-by-4.0
tags:
- concept-bottleneck-models
- phishing-detection
- nlp
---

# NWeak/emails-user-study

Our model's computed results for the 1000-email user-study set (see
`emails/scripts/train_concept_extractor2.ipynb` in
[user-study-CBMs](https://github.com/debryu/user-study-CBMs)): per email,
the 6 ground-truth concepts (`c1..6_gt`), the 6 predicted concept activations
(`c1..6_pred`, continuous -- tanh of the concept-SVM decision function), the
predicted task label (`task_pred`), and the ground-truth label (`task_gt`).

Concept order: Fear+Authority, Urgency, Curiosity, Neutral, Reply, Open attachment.

Also includes `Subject`/`Body`/`Sender` for readability. Join on
`Original email No.`.

## Source dataset and attribution

The emails themselves are **not ours**. They come from the
**PhishingSpamDataSet** of Toth, Bisztray and Dubniczky, released under
CC-BY-4.0 and redistributed here under that licence:

> Rebeka Toth, Tamas Bisztray, Richard A. Dubniczky.
> *Constructing and Benchmarking: a Labeled Email Dataset for Text-Based
> Phishing and Spam Detection Framework.*
> [arXiv:2511.21448](https://arxiv.org/abs/2511.21448) |
> [github.com/DataPhish/PhishingSpamDataSet](https://github.com/DataPhish/PhishingSpamDataSet)

Theirs: `Subject`, `Body`, `Sender`, `URL(s)`, `Type`, `Created by`, `Source`,
`Year`, and the `LLM detected emotion` / `LLM detected motivation`
annotations. Ours: the filtering and split, the embeddings, the 6-concept
merge, and the trained models' outputs.

Please cite Toth et al. alongside our paper if you use this data.

**Note:** `Sender` and `Body` are real email content carried over from the
source dataset.

## Citation

Produced for **[Are Concept Bottleneck Models Effective as Decision-Support
Systems?](https://arxiv.org/abs/2608.25581)** (arXiv:2608.25581) -- Bogani,
Debole, Marconato, Pugnana, Tentori, Passerini. Code:
[github.com/debryu/user-study-CBMs](https://github.com/debryu/user-study-CBMs).