Modelpublic

arun-gharami/lead-ai-review-sentinel

sourceHugging Faceapache-2.0updated 5mo agoView on Hugging Face

0likes

Model Card

Lead.AI Review Sentinel

Model repo target: arun-gharami/lead-ai-review-sentinel

Lead.AI Review Sentinel analyzes customer reviews/messages and flags risky reputation moments before escalation.

Service message: "Detect unhappy customers before they leave bad reviews."

Business Output

—Positive
—Neutral
—Negative
—Urgent Complaint

Along with:

—urgency_score (0-100)
—suggested_reply for support/reputation teams

Features

—review_text
—rating
—response_time_hours
—customer_type
—issue_category

Project Files

—data/data.csv - synthetic Kaggle-ready dataset
—dataset/README.md - dataset card
—train_model.py - hybrid text + tabular classification pipeline
—model/model.joblib - serialized model artifact (generated after training)
—model/metrics.json - evaluation report (generated after training)
—app.py - Gradio demo for live analysis + CSV upload
—sample_api_usage.py - API integration example
—push_to_huggingface.py - publish project to Hugging Face model repo

Local Training

bash

pip install -r requirements.txt
python train_model.py
python app.py

Example Payload

json

{
  "review_text": "Billing error was not fixed after multiple messages.",
  "rating": 1,
  "response_time_hours": 72,
  "customer_type": "returning",
  "issue_category": "billing"
}

Evaluation Snapshot

—Accuracy: 1.0000 (synthetic dataset benchmark)
—Train/Test split: 80/20 stratified
—Metrics artifact: model/metrics.json
—This is a synthetic benchmark. Real-world accuracy may vary.

Pricing

—Starter: $19/month
—Business: $79/month with monitoring + auto-reply suggestions
—Custom AI Setup: $499-$1,500 one-time

Publish to Hugging Face

bash

export HF_TOKEN=hf_xxx
python push_to_huggingface.py