Enterprises have been trying different ways to automate business processes over the last many decades but this has become a very high priority task with the onset of the COVID-19 pandemic. One of the critical components of automation is NLP (Natural Language Processing). With more and more companies opting for automating their internal processes, using virtual interfaces, understanding and mimicking human language and its underlying elements like semantics, phonetics, tonality and even sentiments becomes a critical success factor.
From a linguist’s point of view, NLP is nothing but breaking down language in order for the computer to understand and process it better. It helps eliminate the ambiguity that complex human language poses. A sentence goes through the following stages of analysis.
**picture source – Coursera NLP
Morphology
The study of words, how they are formed, and their relationship to other words in the same language. It analyzes the structure of words and parts of words such as stems, root words, prefixes, and suffixes.
Syntax
Syntax is the order or arrangement of words and phrases to form proper sentences. The most basic syntax follows a subject + verb + direct object formula.
Semantics
Semantics is the study and analysis of how language is used figuratively and literally to produce meaning.
Pragmatics
Pragmatics refers to how words are used in a practical sense. For example, words that attempt to explain abstract concepts-freedom, beauty-have no meaning in and of themselves. Instead, someone who looks at pragmatics would attempt to understand how they are being used in a given, concrete, practical situation.
Most of us who use a digital device would have encountered chatbots in getting some of their stuff done.
Simple chatbots can be programmed with a basic set of rules (“if the user says X, you say Y”); more advanced chatbots or “cognitive agents” use deep learning to self learn from earlier conversations. Language is how humans naturally communicate, so computer interfaces that can understand natural language are more adept at completing tasks without human intervention.
Let’s take as an example, the HR department of an organisation. They have a team of HR professionals who happen to answer the day to day queries of their employees to the tunes of around 200 per day. The queries range from leaves, attendance, income tax, etc. The organisation, to reduce the load of their employees from these mundane tasks, decided to deploy a virtual agent. The virtual agent was designed to behave similarly to human agents, and used ML (machine learning) to attend to most of the queries without needing the help of their employees.
This is just one example and there are many such use cases wherein bots, based on NLP and ML are taking over tasks that earlier required an army of people to do. This not only adds to cost savings for the organisation, but also increases efficiency and most importantly customer satisfaction.
This blog aims at presenting a basic understanding of the concepts that form the basis to language and automation aka intelligent chatbots/virtual agents. The next would aim at diving a little more deeper in how language processing works in automation.
Reference:
https://venturebeat.com/2021/12/08/natural-language-processing-is-shaping-intelligent-automation/
