Airports can automate passenger communications via social media and messaging channels using AirChat, a persona-based 2-way engagement platform. Through webchat, WhatsApp, Messenger, and various other channels, passengers can ask questions and receive immediate responses. Using the channels of their choice, passengers can register for flights and receive real-time flight information based on their persona.
The AI technical part:
NLP (Natural Language Processing) is a branch of artificial intelligence (AI) that focuses on the interaction between computers and human language. It involves analyzing, understanding, and generating natural language text or speech. NLP techniques enable computers to comprehend and respond to human language more meaningfully.
On the other hand, ChatGPT is an advanced language model developed by OpenAI that utilizes NLP techniques. It is trained on a vast amount of text data and can generate human-like responses given a prompt. ChatGPT is designed to engage in interactive conversations, making it particularly useful for chat-based applications.
While NLP encompasses a broader set of technologies and techniques for language processing, ChatGPT is a specific implementation of an NLP model that excels in conversational contexts. ChatGPT leverages NLP to understand and generate responses in a more conversational, human-like manner.
NLP provides foundational techniques and methodologies for processing human language. ChatGPT is an application of NLP focused on generating meaningful and coherent responses in conversational settings.
Within the AirChat technology stack, there are benefits to both NLP and ChatGPT and other similar Large Language Models (LLMs). With NLP, users/airports can decide the training phases and responses to content and conversations. In contrast, ChatGPT and other LLMs require less user/airport effort, reducing control over content.
We’ve blended the two technologies so airports can decide when NLP is the best placed to respond to the passengers’ query and when ChatGPT is best placed to respond. By doing so this reduces the effort required from the airport but also means that on implementation the AirChat platform comes “trained” on vast amounts of data.