Hugging Face's logo

Datasets:
AskFirstAI
/
AskFirst-AI-Contextual-Dialogues-Sample

Tasks:
Text Generation
Modalities:
Image
Text
Languages:
English
Size:
< 1K
Tags:
rag-analytics
long-form-generation
multimodal-prompts
context-in-context
License:
Dataset card Data Studio Files
xet
 Community
1
AskFirst-AI-Contextual-Dialogues-Sample / README.md
AskFirstAI's picture
AskFirstAI
Update README.md
af843bd verified
preview code
|
Raw History Blame Contribute Delete
4.49 kB
metadata 
license: cc-by-nc-4.0
task_categories:
  - text-generation
  - conversational
language:
  - en
tags:
  - rag-analytics
  - long-form-generation
  - multimodal-prompts
  - context-in-context
size_categories:
  - 100k<n<1M

Verified Long-Form Articles & Contextual Dialogues Dataset (Premium Sample)

1. DATASET OVERVIEW

This repository contains a premium production sample of the Content-in-Context dataset managed by Ask First AI, Inc. This multi-layered database is purpose-built for institutional Large Language Model (LLM) fine-tuning, advanced Retrieval-Augmented Generation (RAG) ground-truth validation, multi-turn conversational agent reinforcement, and cross-media text-to-image/video alignment.

  • Total Production Volume: 108,359+ full records (dynamically ingested via real-time backend SDK streaming).
  • Delivery Engine: Secured via Google BigQuery Analytics Hub Data Clean Rooms.
  • Licensing Model: Commercial Enterprise Licenses are handled directly via corporate Proforma Invoices or Private Cloud Offers.

2. PRODUCTION DATA SCHEMA SPECIFICATION

Every row inside this database maps an entire synchronized user-to-knowledge lifecycle containing 16 precision fields:

  • sync_timestamp (TIMESTAMP): Canonical UTC log insertion window.
  • session_id (STRING): Anonymized UUID linking multi-turn interaction tracks within a single user journey.
  • user_question (STRING): Clean human query. 100% anonymized at source (all corporate PII, emails, and phone numbers are redacted).
  • ai_answer (STRING): High-fidelity immediate assistant responses.
  • topic (STRING): Categorical domain labels (science, business, entertainment, legal, lifestyle, healthcare, general_chat).
  • article_title (STRING): SEO-optimized headline for generated long-form materials.
  • article_content (STRING): Expert-level verified long-form article body text (500–1,500 words) styled with clean Markdown layouts (#, ##, code blocks, lists). Left NULL for general chat reactions.
  • idea_output (JSON/STRING): Programmatic JSON strings outlining core conceptual advice or creative directions.
  • analyzer_output (JSON/STRING): High-value JSON maps holding search volume metrics, competition values, trend directions, and key forecasting drivers.
  • image_prompt (STRING): Instruction-compliant English prompts automatically compiled for text-to-image engines.
  • image_url (STRING): Secured qualified URI pointing to the visual ground truth asset inside Google Cloud Storage.
  • video_prompt (STRING): Production voiceover scripts and target cinematic camera cues for text-to-video pipelines.
  • video_url (STRING): Secure Google Cloud Storage URI linking the completed short-form video asset (.mp4).
  • is_super_data (BOOLEAN): Quick-filter flag signaling the unified presence of complex text, full multimedia assets, and analytical context profiles.
  • is_auto_verified (BOOLEAN): Safety filter and fact-checking pass validation flag.
  • has_multimedia (BOOLEAN): Instant boolean filtering anchor for attached multimedia files.

3. ENTERPRISE COMMERCIAL LICENSING & SUBSCRIPTIONS

While this sample is provided under the restrictive Creative Commons Non-Commercial (CC BY-NC 4.0) license for evaluation purposes, our production database is leased to institutional clients via secure BigQuery Shared Exchange Views under the following tiered commercial schedules:

  1. Historical Production Archive (One-time Bulk Purchase): €5,500
    Immediate uncapped read access to the complete baseline database of 108k+ production records.
  2. Small Batch Continuous Ingestion Subscription: €800 / month
    Automated streaming delivery up to 5,000 new verified rows per month.
  3. Enterprise Ingestion Subscription: €2,500 / month
    Automated streaming delivery up to 20,000 new verified rows per month.
  4. Unlimited Real-Time Ingestion Stream: €4,500 / month
    Uncapped, live integration streaming directly from our Next.js SDK pipeline into your model training architecture.

4. CONTACT & DATA ROOM ENQUIRIES

To request a customized data slice (e.g., restricted strictly to science or legal metadata), execute a corporate evaluation agreement, or arrange a private Pricing Schedule Offer, please contact our data administration department directly:

  • Corporate Infrastructure Hub: Ask First AI, Inc.
  • Primary Communications Desk: boss@askfirst.ai