Skip to main content Skip to footer
  • /content/cognizant-dot-com/us/en/glossary

No Results.

Did you mean...

Or try searching another term.

Neural network

What is a neural network?

A neural network is a methodology or set of algorithms that functions similarly to a human brain. It applies deep learning techniques to recognize patterns and draws conclusions without human intervention.

Neural networks, a type of machine learning, learn and refine results over time. They are capable of organically learning and modeling complex, non-linear relationships. They can also find shortcuts, which is highly valuable in big data analysis. Neural networks can infer relationships and self-repair when data is missing or error conditions occur.

What are the business benefits of a neural network?

A deep neural network offers multiple business benefits. For example, it can:

  • Detect fraud. Across multiple industries, neural networking helps prevent fraud by detecting and sending alerts on fraudulent schemes.
  • Enhance customer relationships. Businesses can use neural networks to better identify customer segments, target their marketing and sales efforts and determine why customers may be choosing their competition.
  • Refine marketing initiatives. Evolutionary neural network applications can be used to create segment-specific marketing campaign approaches. For instance:
    • In retail, they can make forecasts more accurate, providing a better picture of which products were purchased on a particular day, how many times and what combinations of products were purchased most often.
    • In finance, they can provide more accurate exchange and stock rate predictions, and enable banks to offer loans based on statistical data collected.
    • On manufacturing floors, they can analyze data from machinery, sensors, beacons and cameras to monitor and optimize plant processes.
    • In insurance, they enable insurers to segment their customers for marketing, pricing and risk purposes.
Neural network featured content

Back to glossary