Skip to content

KEITH-GJINO/Market-Intelligence

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 

Repository files navigation

Headline Sentiment Intelligence (Zero-Shot NLP)

Executive Summary

This project demonstrates a data-first approach to market sentiment. By leveraging Zero-Shot Classification (BART-large-mnli), this tool bypasses the high error rates of legacy, dictionary-based analyzers (like TextBlob) in favor of deep semantic context.

Why This Matters for 2026 Marketing

Traditional SEO is dead. Founders must now optimize for GEO (Generative Engine Optimization). This tool serves as a foundational component for monitoring how AI models perceive brand headlines and market trends in real-time.

High-Level Capabilities

  • Zero-Shot Logic: Classifies headlines into any category without pre-training or hard-coded keyword lists.
  • Confidence Scoring: Outputs a mathematical probability for every label, allowing for high-precision data analysis.
  • Tech Stack: Python 3.12, Hugging Face Transformers, BeautifulSoup4.

Deployment

This repository is optimized for quick deployment in virtual environments to ensure dependency isolation and system hygiene.

Sample Output The following data was generated during a live execution of the analyzer. Note the confidence scores, which highlight the model's ability to handle linguistic nuance in real-time headlines.

Classification complete in 32.0s (1.6s per headline)

                      HEADLINE SENTIMENT ANALYSIS (Zero-Shot)                          
                          Model: facebook/bart-large-mnli                              
                   Source: TechCrunch | March 04, 2026 03:54 AM        
+------+------------------------------------------------------+------------+--------------+
|    # | Headline                                             | Sentiment  |   Confidence |
+------+------------------------------------------------------+------------+--------------+
|    1 | The hidden cost of early hires with Yuri Sagalov,... | Negative   |        71.2% |
|    2 | More from TechCrunch                                 | Positive   |        60.7% |
|    3 | Why AI startups are selling the same equity at tw... | Neutral    |        41.1% |
|    4 | Claude Code rolls out a voice mode capability        | Neutral    |        41.0% |
|    5 | A suite of government hacking tools targeting iPh... | Negative   |        68.3% |
|    6 | Alibaba’s Qwen tech lead steps down after major A... | Negative   |        47.0% |
|    7 | Hacked traffic cams and hijacked TVs: How cyber o... | Negative   |        60.9% |
|    8 | X says it will suspend creators from revenue-shar... | Negative   |        89.2% |
|    9 | The candidate that Silicon Valley built is now th... | Negative   |        81.8% |
|   10 | Apple unveils new MacBook Air and MacBook Pro wit... | Positive   |        50.5% |
|   11 | Just three companies dominated the $189B in VC in... | Positive   |        44.1% |
|   12 | TikTok down for some in US, thanks to second Orac... | Negative   |        85.5% |
|   13 | AI companies are spending millions to thwart this... | Negative   |        74.9% |
|   14 | ChatGPT’s new GPT-5.3 Instant model will stop tel... | Positive   |        35.5% |
|   15 | Anduril aims at $60 billion valuation in new fund... | Positive   |        85.3% |
|   16 | Android users can now share tracker tag info with... | Positive   |        42.2% |
|   17 | ChatGPT uninstalls surged by 295% after DoD deal     | Positive   |        78.6% |
|   18 | MyFitnessPal has acquired Cal AI, the viral calor... | Negative   |        44.3% |
|   19 | Anthropic’s Claude reports widespread outage         | Negative   |        83.7% |
|   20 | The trap Anthropic built for itself                  | Negative   |        92.1% |
+------+------------------------------------------------------+------------+--------------+

About

This project utilizes Zero-Shot NLP (BART-large-mnli) to bypass flawed, dictionary-based sentiment analysis in favor of true semantic context. Designed for high-speed market intelligence and GEO (Generative Engine Optimization) strategy.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages