post Details

Why Natural Language Processing?

The ultimate test developed for deciding if a machine is capable of achieving human behavior is the Turing test. In this test, persons ask another person or machine questions. In the end, the human is asked to decide if on the other side there is a person or a machine. One way of achieving communication between persons is by the exchange of texts.
To solve such tasks computers need to understand texts. The branch of computer science focused on developing such approaches is NLP.

What is NLP?

Natural language processing is the subfield of Artificial Intelligence that is focused on information extraction and analysis of texts. Its roots date as early as the 1950s. It consists of multiple fields: 

• Document and word representations
• Question Answering
• Sentiment analysis 
• Text generating 
• Keyword extraction and generation
• Terminology extraction
• Entity resolution and linking
• Topic modeling
•  

Document and word representations

  In order to process data computers need it to be represented by numbers. This field of NLP explores the approaches of representing texts and words as numbers (or an array of them) in the computer. The naive approaches include representing texts by the count of occurrence of every word in it. This approach is called Bag of Word since for every document it counts the number of words appearing in that document. More advanced approaches include neural networks and try to learn words by their context.

Question Answering

The ultimate tool for tackling the Turing test is a proper question answering algorithm. The field of NLP developed quite some research in this subfield.

Sentiment Analysis

How one tweet will make the followers feel? How will the mail you just sent to the audience going to affect the partners? The sentiment and emotion analysis of the textual data is all about emotions and sentiments in texts. Significant developments have made algorithms that beat the human ability for solving such tasks. 

Text Generation

In many cases, you need automatically created banners, texts, flyers, and so on.

At Quant Coding we have experience working with NLP and Machine Learning, check out our services section, or contact us to arrange a consultation.

This is part ½ of this series on NLP, written by Bosko Koloski, a valued team member and an expert in NLP with credentials from the Josef Stephan Institute in Slovenia, come back for part two of this series to find out more about the techniques of Natural Language Processing.