DEV Community: SHIVASHIGA A.M

Rural - X Smart Farmer Assistance & Government Connected AgriTech Platform

SHIVASHIGA A.M — Sat, 27 Dec 2025 11:51:51 +0000

Agriculture is not just an occupation in India—it’s a lifeline. Yet farmers continue to struggle with fake fertilizers, unpredictable weather, crop diseases, lack of guidance, and slow government communication. What farmers need today is not just another app…
They need a digital companion. A protector. A guide. A community.

That’s exactly what we’re building.
Presenting ― RuralX.

🚀 What is RuralX?

RuralX is a multilingual, AI-powered smart agriculture ecosystem built to support farmers, enhance decision-making, prevent crop loss, and bridge the gap between farmers and government authorities.

It’s not just a tool…
It’s an ecosystem focused entirely on farmer safety, intelligence, and empowerment.

🎯 Why RuralX?
🌱 Real Problems We Are Solving

Fake fertilizers & pesticides destroying crops

Wrong fertilizer ratio causing crop burn

Crop diseases spreading silently

Poor access to government updates

No unified digital system for rural farmers

Lack of structured farming intelligence

No support during disasters like floods & cyclones

RuralX steps in as a digital shield + smart advisor + community connector.

🌍 Multilingual from the First Click

When RuralX opens, the user is greeted with a language selection popup.
Farmers can choose:
Hindi
Tamil
English...

Because technology should never be limited by language.

👨‍🌾 How RuralX Works for Farmers

Farmers sign up with:

Name

Mobile number + OTP

Location

Based on their district, they are automatically added to a community group, similar to a broadcast system.
Examples:

Chennai Community

Madurai Community

Trichy Community

This becomes their trusted communication channel for:

Weather alerts

Government schemes

Fake fertilizer warnings

Agriculture guidance

🏛️ Government Admin Portal

Government officials act as Admins.
They can:

Broadcast alerts

Handle farmer complaints

Validate issues

Monitor region-wise agriculture health

This ensures authentic, verified, and impactful communication.

🌟 Core Features of RuralX
🤖 1️⃣ AI Crop Disease Prediction

This isn’t just disease prediction. It’s smarter.

Farmer uploads leaf image

AI analyzes it

If confidence > 80% → actionable cure is provided

If confidence < 80% → asks for re-upload

If still low → Automatically alerts government admin

Behind the scenes:

Meta-learning techniques

Better model adaptability

Real-world intelligence

Optionally using Gemini / LLM support for cure explanation

This ensures safe, reliable, trustworthy disease detection.

🧪 2️⃣ Fertilizer Ratio + Burn Risk Advisor

Farmers input:

Crop type

Soil type

Land area

System outputs:

Exact fertilizer quantity

Burn risk safety indicator

Suggested verified fertilizer shops

Result?
➡ Safer farming
➡ Better productivity
➡ Reduced crop loss

🧠 3️⃣ “Soil Memory” & Weather Awareness

RuralX remembers:

What crops the farmer previously grew

What soil combinations they used

Then suggests:

Best crop recommendations

Ideal farming windows

Weather alerts automatically reach the farmer’s district community.

📅 4️⃣ Growth Calendar

A structured crop growth journey with reminders and guidance.
Think of it like a personal assistant for crops.

📡 5️⃣ Smart Offline Behavior

If network drops:

A popup notifies the farmer

Previously stored values are still shown

Because RuralX is built for real rural conditions.

💧 6️⃣ Irrigation Simulation

Farmers struggle with sprinkler planning.
RuralX helps by:

Simulating field sprinkler setup

Showing blueprint-style layout

Suggesting exact number & placement

Enhancing water distribution efficiency

This means:
➡ Better crop health
➡ Reduced water waste

🌊 7️⃣ Disaster Management System

During:

Floods

Cyclones

Heavy rainfall

RuralX:

Suggests drainage planning

Helps farmers edit layouts

Uses AI guidance for optimal protection

A life saver when nature turns unpredictable.

🏗️ System Architecture (Simple View)

Farmer → Next.js Frontend
→ Flask Backend
→ MongoDB Database
→ AI Intelligence Layer
→ Insights / Alerts / Support

Admin Portal → Analytics + Broadcast → Farmer Communities

💥 Impact

RuralX can:
✔ Reduce crop loss
✔ Prevent financial damage
✔ Increase agricultural safety
✔ Strengthen Government–Farmer trust
✔ Modernize rural India

🌾 RuralX — More Than a Project

RuralX isn’t just software.
It’s a mission to empower farmers.
A step toward intelligent, safe, and sustainable agriculture.

The future of farming shouldn’t be uncertain.
With RuralX, it doesn’t have to be.

Defect Detection in FFF 3D printing using Meta-Learning [Prototypical Networks]

SHIVASHIGA A.M — Thu, 31 Jul 2025 05:29:23 +0000

Teaching AI to Spot 3D Printing Mistakes (Just Like a Human)
Ever printed a 3D part only to find it cracked, warped, or stringy? It’s like watching your beautiful idea melt into disappointment.
Well, what if we could teach a machine to spot those flaws before we waste time and material?

That’s exactly what I set out to solve.✌️

🚀 The Idea: Catching 3D Printing Defects Early
3D printing is magical — it brings digital dreams to life. But it’s also tricky.
Things like layer shifting, stringing, or warping can sneak in and ruin the whole print. In manufacturing, that could mean hours of downtime or thousands in material loss.

So…
Can we make an AI that can detect those mistakes just by looking at the print?
And more interestingly…
Can it learn to do that from just a handful of examples — like a human does?

🧠 Meet Prototypical Networks — The Human Way to Learn
Imagine you’ve seen 2 or 3 broken parts before.
Now when someone shows you a new one, your brain just knows something’s wrong.
That’s exactly how Prototypical Networks work.

Instead of training a model on thousands of examples, this method learns the essence (or "prototype") of each defect from just a few samples.

Here’s how it works:
How the AI Learns??

Support Set: A few labeled images per defect type (Normal, Warping, Cracking, etc.)
Query Image: A new, unlabeled image the model has never seen before
Feature Extraction: A CNN converts images into feature vectors
Prototype Formation: It averages the features of known examples per class
Prediction: The query image is compared with each prototype; the closest match wins!

💣Boom — your defect is classified.

What Kinds of Defects Are We Spotting?

These are common enemies of any 3D printing hobbyist or professional.

📊 But Does It Work?
I tested this system on real-world FDM printed part images.
Even with very few training examples per defect, the model achieved promising accuracy and generalization, thanks to the prototype-based learning.

We evaluated it using:

✅ Accuracy: Overall correctness [0.98]

📈 F1-Score: Balanced look at precision & recall

🔁 Confusion Matrix: Visual check on prediction mix-ups

And yes — the AI really did start thinking like a human in a way.

🌟 Why This Matters
Most factory floors can’t afford to collect thousands of examples for every new defect.
But using few-shot learning like this makes AI more practical, scalable, and smart — even in low-data environments.

This can lead to:

Fewer failed prints
Less material waste
Smoother workflows
Happier makers 😄

💡 What’s Next?
I’m exploring ways to:

Integrate this with a live camera feed
Use multi-modal inputs (temperature, sound, etc.)
Deploy it on Raspberry Pi or Jetson Nano near the printer

Imagine: A smart 3D printer that taps you on the shoulder and says:

“Hey, something looks off. Wanna check this before we go further?”

❤️ Final Thoughts
This project blends computer vision, few-shot learning, and a real-world problem that makers and industries alike face every day.

It’s not just about pixels and prototypes — it’s about making machines that understand, adapt, and help us build better things.

If you’ve ever yelled at your 3D printer (or wanted to), you know why this matters.

Let’s bring more brains to the bench — one prototype at a time.

🔗 Wanna See the Code?
You can find the full project and code on GitHub.
If you’re working on a similar problem or want to collaborate, feel free to reach out!

DataPlan Recommender: ML-Powered Personalized Internet Plan Suggestions

SHIVASHIGA A.M — Tue, 29 Jul 2025 05:02:58 +0000

In today’s hyperconnected world, choosing the right data plan can be overwhelming. Most people settle for what’s available without knowing if it’s cost-efficient or tailored to their needs.

That’s where DataPlan Recommender steps in — a machine learning-powered platform that personalizes data plan suggestions based on individual usage behavior and demographics.

🧠 What is DataPlan Recommender?

DataPlan Recommender is a smart system that leverages CatBoost and XGBoost machine learning algorithms to recommend the most suitable and cost-effective data plans. It analyzes user data including demographics, occupation, location, and usage patterns to ensure that users get personalized and optimized suggestions.

The core goal? Reduce unnecessary expenses and ensure uninterrupted, value-driven internet access.

🗂 Dataset Overview

The project uses a rich dataset with diverse attributes that mirror real-world telecom usage scenarios:

👤 Personal Info: Name, Age, Mobile Number, Password

💼 Occupation: IT Professionals, Doctors, Teachers, etc.

📍 Location: Users from various cities

📊 Usage: Daily & Monthly Internet Usage (GB/MB)

📱 Devices: Smartphone, Tablet, Smart TVs

🌐 Network Preferences: 4G/5G, SIM Type (Airtel, VI, BSNL)

📦 Package Details: Data Limit, Package Cost, and Network Plan

This data helps identify usage trends and predict the most fitting data plan for each user.

Methodology:

The methodology involves a structured machine learning pipeline:

Data Preprocessing – Clean, normalize, and format user data
Feature Engineering – Extract insights like usage trends, cost per GB
Model Training – Use CatBoost/XGBoost to train prediction models
Backend Integration – Flask APIs to serve predictions
Frontend Interaction – Responsive UI with real-time suggestions

📊 The system learns from historical data to uncover the best match between user profile and available data plans.

💻 Tech Stack Here’s what powers the application:

📈 Results The application delivers effective and insightful outcomes:

✅ Prediction Accuracy: High accuracy (0.98) in predicting user needs using advanced models

🙌 User Engagement: Intuitive front-end helps users compare and choose plans easily

💸 Optimized Costs: Suggests cost-efficient plans based on actual usage

📡 Telco Insights: Provides feedback to network providers for better pricing models

🖼 Snapshots Here are some glimpses of the platform in action:

🎥 Demo See the recommender system in action!

👉 Watch Demo Video

Github repo link:

👉 refer here

🚀 What’s Next?

I'm working on:

Adding more user behavioral data like app usage patterns
Supporting real-time plan updates via telecom APIs
Creating a mobile-first version of the platform
Integrating with billing portals for one-click plan purchases

💬 Final Thoughts The DataPlan Recommender showcases how AI/ML can transform something as mundane as choosing a mobile data plan into a smart, personalized, and cost-efficient process. Whether you’re a student, a working professional, or a telco provider, this tool offers something valuable.