Tgram3D/education-forum-jfk-dataset

Education Forum JFK Assassination Debate Dataset

Overview

This dataset contains a structured archive of public discussion threads centered on the JFK Assassination Debate section of the Education Forum.

Beginning with Version 2.0, the dataset also includes selected discussion forums from other sections of the Education Forum while retaining the original dataset name for continuity and discoverability.

The Education Forum spans more than two decades of discussion and includes conversations among historians, researchers, authors, journalists, educators, and members of the broader research community.

The dataset has been curated into structured JSON and JSONL formats for historical research, natural language processing, information retrieval, digital humanities, and long-term archival purposes.

Dataset Statistics

• —Version: 2.0.1
• —Forums: 46
• —Threads: 25,187
• —Comments: 513,116
• —Time Span: December 2003 – June 18, 2026
• —Source Forum: Education Forum – JFK Assassination Debate

Repository Contents

README.md
schema.json
metadata.json
utils.py

threads_jsonl/
├── forums.jsonl
├── threads.jsonl
└── comments.jsonl

EducationForum_ThreadArchive.zip

Included Forums

The dataset is centered on the JFK Assassination Debate forum while also incorporating selected discussion forums from other sections of the Education Forum.

Each discussion thread belongs to a forum identified by forum_id. The forums.jsonl file provides metadata for every included forum, allowing the dataset to be filtered or analyzed by individual forum or category.

Dataset Formats

JSONL Dataset (Recommended)

The threads_jsonl/ directory contains the normalized dataset.

forums.jsonl

One JSON object per forum.

Each record includes:

• —forum_id
• —title
• —url
• —category
• —thread_count
• —comment_count

threads.jsonl

One JSON object per discussion thread.

Each record includes:

• —thread_id
• —forum_id
• —title
• —url
• —author
• —author_url
• —datetime
• —date_text
• —forum_page
• —comment_count
• —scrape_error

comments.jsonl

One JSON object per forum comment.

Each comment references its parent discussion thread using the thread_id field.

Each record includes:

• —comment_id
• —thread_id
• —author
• —author_url
• —user_id
• —datetime
• —date_text
• —text
• —content_html
• —comment_index
• —thread_page
• —index_on_page
• —source_url

The content_html field preserves the original HTML structure, including:

• —quotations
• —hyperlinks
• —embedded media
• —formatting
• —block quotes
• —lists

The text field provides a plain-text representation suitable for searching, indexing, and NLP applications.

Thread Archive

The archive

EducationForum_ThreadArchive.zip

contains the original archival representation of every scraped discussion thread.

Threads are organized by Education Forum category and forum, preserving the original forum hierarchy.

Each discussion thread is stored as an individual JSON document containing:

• —complete thread metadata
• —chronological comment ordering
• —author information
• —timestamps
• —hyperlinks
• —preserved HTML
• —embedded media references

This format is intended for historians, researchers, archivists, and users wishing to browse complete forum discussions.

Relationship Between Files

forums.jsonl
      │
      └── forum_id
              │
              ▼
threads.jsonl
      │
      └── thread_id
              │
              ▼
comments.jsonl

Each forum contains one or more discussion threads.

Each thread contains one or more comments.

Comments can be joined to their parent forum through thread_id.

Helper Utilities

The included utils.py module provides helper functions for loading the JSONL dataset and retrieving complete discussion threads.

Example:

from utils import get_thread_by_id

thread, comments = get_thread_by_id("32102")

Potential Uses

This dataset may be useful for:

• —historical research
• —digital humanities
• —information retrieval
• —retrieval-augmented generation (RAG)
• —natural language processing
• —large language model evaluation
• —forum search engines
• —quote network analysis
• —citation extraction
• —topic modeling
• —temporal analysis
• —social network analysis
• —text mining

Source

Education Forum

Primary collection:

https://educationforum.ipbhost.com/forum/126-jfk-assassination-debate/

Additional Education Forum discussion forums are included beginning with Version 2.0.

Notes

• —All timestamps are stored in ISO 8601 format where available.
• —Identifiers (forum_id, thread_id, comment_id, and user_id) are stored as strings for consistency across platforms.
• —Rich HTML formatting is preserved alongside plain-text content.
• —Some guest posts do not have an associated member profile and therefore have a null author_url.
• —The thread archive preserves the original forum hierarchy by category and forum.
• —The dataset contains publicly accessible forum content only.

License

This repository contains publicly accessible forum content archived for research, educational, historical, and analytical purposes.

Copyright in individual forum posts remains with their respective authors and/or the Education Forum.

The dataset structure, metadata, schema, and utilities were created by the dataset author. Users are responsible for ensuring that their use of the archived forum content complies with applicable laws and the Education Forum's terms of use.

Citation

If you use this dataset in research or publications, please cite:

Gram, T. (2026). Education Forum JFK Assassination Debate Dataset. Hugging Face.

If you used a specific release, please cite the corresponding version (e.g., v2.0.0) or the associated Hugging Face dataset revision.