metadata
language: en
license: mit
library_name: transformers
tags:
- text-classification
- intent-detection
- b2b
- sales-intelligence
- gtm
- fine-tuned
datasets:
- custom-synthetic
metrics:
- f1
- accuracy
pipeline_tag: text-classification
B2B Intent Signal Classifier
Fine-tuned DistilBERT model that classifies B2B company signals into 6 intent categories for GTM teams.
Model Description
This model detects buyer intent signals from text — news articles, job postings, press releases — and classifies them into actionable categories for sales and marketing teams.
Categories
| Category | Description | Example |
|---|---|---|
hiring_surge |
Aggressive hiring in specific area | "TechCorp hiring 50 ML engineers" |
funding_round |
Investment received | "DataCo closes $50M Series C" |
product_launch |
New product/feature release | "CloudInc launches AI platform" |
leadership_change |
Executive hired/departed | "New CTO joins from Google" |
expansion |
Geographic/market expansion | "Opens Singapore office" |
cost_cutting |
Layoffs, budget cuts | "Reduces workforce by 20%" |
Performance
- F1 Score (Macro): 0.9976
- Accuracy: 0.9976
- Precision: 0.9976
- Recall: 0.9976
Usage
from transformers import pipeline
classifier = pipeline("text-classification", model="SrihariV/b2b-intent-signal-classifier")
result = classifier("Stripe raises $200M Series D led by Sequoia")
# Output: [{'label': 'funding_round', 'score': 0.98}]
Training Details
- Base model: distilbert-base-uncased
- Training data: 5,400+ synthetic B2B signal examples
- Epochs: 4
- Learning rate: 2e-5
- Batch size: 32
Built By
Srihari Venkatesan — GTM Engineer | AI Consultant
License
MIT