Research for industry use cases of Neural Networks.

Sanat Dash
6 min readJun 20, 2021

What is Neural Network?

Neural Networks – an Intuition - KDnuggets

Neural networks are a set of algorithms, they are designed to mimic the human brain, that is designed to recognize patterns. They interpret data through a form of machine perception by labeling or clustering raw input data. Let’s take a moment to consider the human brain. Made up of a network of neurons, the brain is a very complex structure. It’s capable of quickly assessing and understanding the context of numerous different situations. Computers struggle to react to situations in a similar way. Artificial Neural Networks are a way of overcoming this limitation. Sometimes called perceptrons , an Artificial Neural Network is a hardware or software system.

Some networks are a combination of the two. Consisting of a network of layers this system is patterned to replicate the way the neurons in the brain operate. The network comprises an input layer, where data is entered, and an output layer. The output layer is where processed information is presented. Connecting the two is a hidden layer or layers. The hidden layers consist of units that transform input data into useful information for the output layer to present. In addition to replicating the human decision making progress Artificial Neural Networks allow computers to learn. Their structure also allows ANN’s to reliably and quickly identify patterns that are too complex for humans to identify. Artificial Neural Networks also allow us to classify and cluster large amounts of data quickly.

Improving Search Engine Functionality During 2015 Google I/O keynote address in San Francisco, Google revealed they were working on improving their search engine. These improvements are powered by a 30 layer deep Artificial Neural Network. This depth of layers, Google believes, allows the search engine to process complicated searches such as shapes and colors. Using an Artificial Neural Network allows the system to constantly learn and improve. This allows Google to constantly improve its search engine. Within a few months, Google was already noticing improvements in search results. The company reported that its error rate had dropped from 23% down to just 8%. Google’s application shows that neural networks can help to improve search engine functionality. Similar Artificial Neural Networks can be applied to the search feature on many e-commerce websites. This means that many companies can improve their website search engine functionality. This allows customers with only a vague idea of what they want to easily find the perfect item. Amazon has reported sales increases of 29% following improvements to its recommendation systems.

Search engines are incredible. You pop onto your favorite one, type in a few keywords, and magically the search engine has scoured the entire internet to find the most relevant content. Except it’s not magic, of course. It’s algorithms and a lot of ideas from a lot of brilliant people. And we’ve started to wonder: With all the brilliant minds behind them, to what extent are search engines using artificial intelligence?

Thanks to those same search engines, I was able to find some great research explaining how search engines use artificial intelligence.

Search Engines Use Artificial Intelligence For Quality Control

Back in the day, certain SEO “specialists” beat the system with shady practices that we’ve come to know as “black hat SEO techniques.” These include aggressive keyword stuffing, cloaking, invisible text — the list goes on. Of course, this was damaging to search engines because the pages that were at the top of their results weren’t necessarily the highest quality content. Nowadays, they’ve updated their algorithms and use AI to separate the high quality content from the low quality spam. We suspect that as AI progresses, it will completely take over this responsibility and remove the need for human quality raters entirely.

Search Engines Use AI To Create Ranking Algorithms

Not only does artificial intelligence protect search engines from manipulation, but it also helps them with their ranking algorithms. It’s impossible to tell how big of a role AI plays in this, but search engines definitely use artificial intelligence to improve their ranking algorithms. To get a little more technical, this specific area of artificial intelligence involves learning to rank algorithms. Machines are taught to create an optimal list from a set of possible outcomes, learning from each of the variables over time. For example, if one result on a search engine is ranking third but has a higher click through rate than the options above it, the search engine would learn from this anomaly and bump that result to the top.

Search Engines Use NLP and Image Analysis To Understand Search Queries

Search engines are computer applications, but they need to be able to understand human language in order to find users the information they’re looking for.

That’s a textbook application of natural language processing (NLP), a field of AI dedicated to teaching computers to understand our written language. At the very least, both Google and Microsoft are using NLP to understand their users. A University of Washington study looked at Yandex, the world’s fourth largest search engine, and noticed that it has some advanced applications of NLP and machine learning. Yandex developers found that they could take all of their users’ previous searches and use them to optimize future searches. By creating these personalized search results, they’ve increased click through rate by about 10%.

Common Applications of Neural Networks

  1. Handwriting Recognition
  2. Traveling Salesman Problem
  3. Stock Exchange Prediction

Handwriting Recognition

The idea of using feedforward networks to recognize handwritten characters is straightforward. The bitmap pattern of the handwritten character is input, with the correct letter or digit as the desired output. Such programs need the user to train the network by providing the program with their handwritten patterns.

The two common applications of handwriting recognition are:

  • Optical character recognition for data entry.
  • Validation of signatures on a bank cheque.

Traveling Salesman Problem

The traveling salesmen problem refers to the finding the shortest possible path to travel all cities in a given area. We can use Neural Networks to solve this problem.

A neural network algorithm such as a genetic algorithm starts with random orientation of the network, to solve the problem. This algorithm chooses a city in a random manner each time and finds the nearest city. Thus, this process continues several times. After every iteration, the shape of the network changes and network converges to a ring around all the cities.

The used algorithm minimizes the length of rings. In this way, we can estimate the traveling problem.

Stock Exchange Prediction

The prediction accuracy of neural networks has made them useful in making a stock market prediction. For large business companies, making predictions for stock exchange is common. This is by using parameters, such as current trends, political situation, public view, and economists’ advice.

Application of Neural Network in Healthcare

  1. Imaging analytics and diagnostics
  2. Drug discovery and precision medicine
  3. Clinical decision support and predictive analytics

Imaging analytics and diagnostics

One type of deep learning, known as convolutional neural networks (CNNs), is particularly well-suited to analyzing images, such as MRI results or x-rays.

CNNs are designed with the assumption that they will be processing images, according to computer science experts at Stanford University, allowing the networks to operate more efficiently and handle larger images.

As a result, some CNNs are approaching — or even surpassing — the accuracy of human diagnosticians when identifying important features in diagnostic imaging studies.

Drug discovery and precision medicine

Precision medicine and drug discovery are also on the agenda for deep learning developers. Both tasks require processing truly enormous volumes of genomic, clinical, and population-level data with the goal of identifying hitherto unknown associations between genes, pharmaceuticals, and physical environments.

Deep learning is an ideal strategy for researchers and pharmaceutical stakeholders looking to highlight new patterns in these relatively unexplored data sets — especially because many precision medicine researchers don’t yet know exactly what they should be looking for.

Thank you…!!