---
title: "Short note: Neural networks"
subject: "Data Mining"
chapter: 3
key: "semester_7/data_mining:neural-networks-note"
source: https://www.bctengineer.com/q/semester_7/data_mining/neural-networks-note
site: BCT Engineering
---

# Short note: Neural networks

Data Mining · Chapter 3: Classification

- Asked in: 2063, 3 marks
- Video: [Lec-35: Introduction to Neural Networks with Example in HINDI | Artificial Intelligence](https://www.youtube.com/watch?v=EYeF2e2IKEo) — Gate Smashers
- Answer checked by Claude Opus 5.5 (high effort) on 2026-10-14
- Page: https://www.bctengineer.com/q/semester_7/data_mining/neural-networks-note
- Whole chapter: https://www.bctengineer.com/md/semester_7/data_mining/chapter-3

## Question

Write a short note on Neural Networks.

## Answer

An **Artificial Neural Network (ANN)** is a computing model inspired by the human brain: a network of simple, connected processing units (**neurons**) that learns from examples. It is used for classification, prediction and pattern recognition.

#### Structure
- **Input layer**: one node per attribute.
- **Hidden layer(s)**: process the signals.
- **Output layer**: gives the class.
- Each link has a **weight**. A neuron computes the weighted sum of inputs plus a bias and passes it through an **activation function** (step, sigmoid): y = f(Σ wᵢxᵢ + b).

```text
 x1 --w1--\
 x2 --w2---> [ Σ + f ] --> output y
 x3 --w3--/
```

#### Learning
Weights start random and are adjusted by training on labelled tuples. The most common algorithm is **backpropagation**: the output error is propagated backwards and the weights are updated by gradient descent until the error is small.

#### Features
- Can model complex non-linear relations and tolerates noisy data.
- Works on continuous and discrete data and gives good accuracy.
- Weaknesses: long training time, many parameters to tune, and a "black box" model that is hard to interpret.

## Check yourself

The answer's key points as multiple-choice questions, with the correct option and why.

1. What does a neuron compute?
   A. The Euclidean distance between inputs
   B. The entropy of its inputs
   C. The majority class of its inputs
   D. A weighted sum of inputs plus bias passed through an activation function
   Answer: D. y = f(sum wi xi + b). (Key point: Neuron computation)

2. What does backpropagation do?
   A. Propagates the output error backwards and updates weights by gradient descent
   B. Prunes unused hidden layers
   C. Sets the weights by counting class frequencies
   D. Splits attributes by information gain
   Answer: A. Weights are adjusted until the error is small. (Key point: Backpropagation)

3. Which is a weakness of neural networks?
   A. They cannot model non-linear relations
   B. Long training time and a black-box model hard to interpret
   C. They cannot tolerate noisy data
   D. They work only on categorical data
   Answer: B. They handle non-linearity and noise but are slow to train and hard to interpret. (Key point: Weaknesses)