Tech Term Decoded: Neural Network

Definition

A neural network, also referred to as artificial neural network (ANN) is a set of algorithms developed to identify patterns and relationships in data via a process that imitates the way the human brain works. A neural network is basically made up of neurons, which are the building blocks similar to brain cells. These neurons receive inputs, process them, and produce an output. They are grouped into different layers: an Input Layer that receives the data, several Hidden Layers that process this data, and an Output Layer that provides the final decision or prediction [1].

For instance, imagine a scenario where the Joint Admissions and Matriculation Board has a candidate screening system with a strict hierarchy. JAMB desk officers process applications below 200 UTME score, senior officers handle candidates below 300 UTME score, and the registrar approves any candidates exceeding 300 UTME score. When students across submit their university applications, they first go to desk officers. Any application over 200 score gets passed to a senior officer, while applications below 200 score stay at the desk level, and so on.

This scenario of candidate screening system of JAMB functions somewhat like a neural network. When students submit their university applications, this is like a neural network's input layer. Each desk officer and senior officer is like a node within the neural network.

 An example of a neural network process [2].

Origin

The origin of neural networks dates back to 1943 when mathematicians Warren McCulloch and Walter Pitts developed a system of circuits that used simple algorithms to mimic the human brain. However, it took another 60 years (around 2010) for their developments to gather momentum as a result of the rise of the Big Data trend, where companies began amassing substantial volumes of data.  

The creation of AlexNet, a unique deep convolutional neural network, by Alex Krizhevsky, Ilya Sutskever, and Geoffrey E. Hinton, marked one of the significant events in the history of neural networks. The AlexNet won the ImageNet challenge, and could classify 1.3 million high-resolution images, achieving an error rate of 15.3%.

This achievement sparked an explosion in deep learning research, and technology companies began to massively invest in classification technologies using AlexNet to create other models for developing realistic problem solutions, such as image detection, image classification, and much more [3].

Context and Usage

When talking about neural networks, there are endless potentials and numerous applications that cuts across various sectors. Some of them are as follows;

• Healthcare: Medical imaging systems use ANN for identifying diseases such as cancer and in diagnosing conditions in general, with accuracy similar to doctors. Furthermore, they predict health risks and suggest personalized treatment plans.
• Social Media: Social media platforms perform tasks such as suggesting friends and relevant content using artificial neural network (ANN) based on analyzing user profiles, interests and interactions. In addition to that, ANN help achieve targeted advertising which ensures users see ads tailored to their preferences.
• Personal Assistants: Virtual assistants like Siri and Alexa use ANNs to process natural language, understand voice commands and respond accordingly. They help manage tasks like setting reminders, making calls and answering queries.
• Customer Support: ANNs drive chatbots and automated customer service systems that process customer queries and provide accurate responses, thereby optimizing customer response processes.
• Finance: By processing large sets of transaction data and spotting anomalies, ANN enable financial sector perform tasks like fraud detection, credit scoring and predicting market trends [4].

Why it Matters

Neural networks can identify and capture the relationships between input and output data that are nonlinear and complex. And by so doing, they enable computers make intelligent decisions with minimum human input. In addition, neural networks can understand unstructured data and make general observations without explicit training. For instance, they can recognize that two different input sentences have a similar meaning:

• How much does it cost to register for JAMB examination?
• What is the current fee for UTME registration? [5]

In Practice

Google Translate is a good example of neural network in practice. You can use Google Translate to automatically translate the text contained in an image. For instance, you could take a picture of a street sign or handwritten note, and Google Translate will scan it and provide a translation. 

See Also

Related AI Models and Architectures:

• RoBERTa: Robustly Optimized BERT Pretraining Approach, an improved transformer language model

Reference

1. Awan, A., A. (2023). What are Neural Networks?
2. Cloudflare. (2025). What is a neural network?
3. Rojewska, K. (2023). What are Neural Networks and What are Their Applications?
4. Geeksforgeeks. (2025). Artificial Neural Networks and its Applications.
5. AWS. (2025). What is a Neural Network?