Bhavin1905/Social-Media-Posts-Dataset-Embeddings-Included-DUCKDB

📊 Social Media Posts Dataset (Embeddings Included)

Dataset Description

This dataset contains social media posts collected for the purpose of natural language analytics and semantic analysis. It is designed to support trend analysis, topic discovery, sentiment inference, and time-based analytics over historical social media data.

The dataset is intended to serve as the data backbone for a natural language analytics system where users can ask questions in plain English and receive data-backed analytical responses.

Dataset Structure

Each row in the dataset represents a single social media post.

Columns

Dataset Size

• —Number of records: ~120,000
• —Time coverage: October 2025
• —Language: English

Embeddings

The embedding column contains precomputed dense vector representations of the post text.

These embeddings enable:

• —Semantic search
• —Topic clustering
• —Meaning-based filtering
• —Complaint and theme discovery

Embeddings were generated once and stored to ensure deterministic, reproducible analytics.

Intended Use

This dataset is intended for:

• —Natural language analytics systems
• —Semantic search and clustering
• —Topic modeling and trend detection
• —Time-series analysis of social media discussions
• —Demonstrations of LLM + classical analytics integration

It is especially suited for:

• —Analytics agent orchestration (e.g., LangGraph)
• —Enterprise-style explainable analytics
• —Historical dataset analysis (non–real-time)

Not Intended For

This dataset does not contain:

• —User identity information
• —Engagement metrics (likes, shares, comments)
• —Platform-specific metadata
• —Geographic data
• —Ground-truth sentiment labels

As a result:

• —Influencer analysis and reach estimation are out of scope
• —Sentiment must be derived from text, not assumed

Notes on Time Awareness

This dataset represents historical data only and does not include real-time updates.

All relative time expressions (e.g., “last week”, “this month”) should be interpreted relative to the most recent timestamp in the dataset, not the current date.

This ensures meaningful analytics over fixed historical windows.

Example Use Cases

• —“What was the sentiment trend in the last available week?”
• —“What topics were people discussing most frequently?”
• —“What are the most common complaint themes?”

Each question can be answered using deterministic analytics derived from this dataset.