---
title: "Neural network; supervised vs unsupervised"
subject: "Artificial Intelligence"
chapter: 6
key: "semester_6/artificial_intelligence:nn-supervised-unsupervised"
source: https://www.bctengineer.com/q/semester_6/artificial_intelligence/nn-supervised-unsupervised
site: BCT Engineering
---

# Neural network; supervised vs unsupervised

Artificial Intelligence · Chapter 6: Machine Learning

- Asked in: 2069 Poush, 6 marks
- Video: [Supervised vs. Unsupervised Learning](https://www.youtube.com/watch?v=W01tIRP_Rqs) — IBM Technology
- Answer checked by Claude Opus 5.5 (high effort) on 2026-10-14
- Page: https://www.bctengineer.com/q/semester_6/artificial_intelligence/nn-supervised-unsupervised
- Whole chapter: https://www.bctengineer.com/md/semester_6/artificial_intelligence/chapter-6

## Question

What is neural network? Differentiate between supervised and unsupervised learning.

## Short answer

A neural network is a computing system of interconnected artificial neurons that learns by adjusting weights from data. Supervised learning uses labeled input-output pairs and a teacher signal to learn a mapping (e.g., back-propagation), while unsupervised learning uses only unlabeled data to discover patterns or clusters without a teacher (e.g., Kohonen SOM).

## Answer

#### Neural network
An **artificial neural network (ANN)** is a computing system made of many interconnected simple processing units (neurons) that work in parallel. Each connection carries a weight; the network learns by adjusting the weights from data. It is inspired by the biological brain and is used for classification, function approximation and pattern recognition.

Structure: input layer, one or more hidden layers, output layer. Each neuron computes y = f(Σ wᵢxᵢ + b).

#### Supervised vs unsupervised learning
| Point | Supervised learning | Unsupervised learning |
|---|---|---|
| Training data | Labelled: input with the desired output (target) | Unlabelled: inputs only |
| Teacher | A teacher/target gives the error signal | No teacher |
| Goal | Learn mapping from input to output | Discover structure, clusters or patterns |
| Feedback | Error = target − actual | None, or internal measure (similarity) |
| Tasks | Classification, regression | Clustering, dimensionality reduction, associations |
| Examples | Perceptron, back-propagation, decision trees | Kohonen SOM, k-means, competitive learning |
| Evaluation | Easy (compare with known labels) | Harder, no ground truth |
| Example | Spam detection from labelled emails | Grouping customers by buying behaviour |

## Check yourself

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

1. What is the key difference in training data?
   A. Supervised uses labelled data with targets; unsupervised uses unlabelled inputs only
   B. Supervised uses unlabelled data; unsupervised uses labels
   C. Both need target outputs
   D. Supervised needs no data
   Answer: A. Supervised learning has a teacher giving the desired output. (Key point: Training data)

2. Which pair is correctly classified?
   A. Kohonen SOM is supervised; back-propagation is unsupervised
   B. Back-propagation is supervised; Kohonen SOM is unsupervised
   C. k-means is supervised; perceptron is unsupervised
   D. Both are unsupervised
   Answer: B. Perceptron and BP use targets; SOM, k-means use none. (Key point: Examples)

3. Which tasks belong to unsupervised learning?
   A. Classification and regression
   B. Spam detection from labelled emails
   C. Clustering and dimensionality reduction
   D. Predicting house price from past sales
   Answer: C. Unsupervised learning discovers structure in the data. (Key point: Tasks)

4. Why is evaluating unsupervised learning harder?
   A. It always takes longer to compute
   B. It uses too many labels
   C. It has no input data
   D. There is no ground truth to compare against
   Answer: D. Supervised results can be compared with known labels. (Key point: Evaluation)

## Related in Artificial Intelligence

- [Short note: supervised vs unsupervised](https://www.bctengineer.com/question.md?key=semester_6%2Fartificial_intelligence%3Asupervised-vs-unsupervised-note) — Artificial Intelligence, chapter 6
- [Neural network; back-propagation](https://www.bctengineer.com/question.md?key=semester_6%2Fartificial_intelligence%3Aneural-network-bp-algorithm) — Artificial Intelligence, chapter 6
- [Role of perceptron in neural network](https://www.bctengineer.com/question.md?key=semester_6%2Fartificial_intelligence%3Aperceptron-role) — Artificial Intelligence, chapter 6, asked 2×
- [Perceptron; neural network as OR gate](https://www.bctengineer.com/question.md?key=semester_6%2Fartificial_intelligence%3Aperceptron-or-gate) — Artificial Intelligence, chapter 6, asked 2×
- [Perceptron; neural network as XOR gate](https://www.bctengineer.com/question.md?key=semester_6%2Fartificial_intelligence%3Aperceptron-xor-gate) — Artificial Intelligence, chapter 6, asked 2×
- [Neural network, back-propagation, perceptron](https://www.bctengineer.com/question.md?key=semester_6%2Fartificial_intelligence%3Ann-bp-perceptron) — Artificial Intelligence, chapter 6

## Related in other subjects

- [Supervised vs unsupervised learning](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Asupervised-vs-unsupervised) — Data Mining, chapter 3
- [Short note: Neural network, perceptron rule](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Aneural-network-perceptron-note) — Data Mining, chapter 3