In the Explained series of blog posts, we break down complex technologies incorporated in our artificial intelligence, Aplysia. In this entry, take a step back to get a better overview of the foundations of HiJiffy’s AI framework.

The origins of our solution

Speaking of foundations, let’s go back to the origins of HiJiffy. Since starting the company in 2016, our primary goal has been to develop an intelligent system that enhances communication between guests and hoteliers, addressing the sector’s needs. Since then, the increase in the use of artificial intelligence in hotels has been found to improve the guest experience significantly by optimising its performance in the delivery of services.

In the last eight years, HiJiffy’s solution has developed into a robust Guest Communications Hub. Even though sometimes it still may be easier, or more immediate, to generally explain our solution as “an AI chatbot”, it is, in fact, just one of the key components of our technological framework. 

From conversational AI to a range of tools and insights available in our Console (the user interface of our solution), our technology incorporates cutting-edge developments in the field to help hoteliers and hotel teams provide more efficient services and understand their guests better. 

Let’s look at the key technologies that enable that to happen. 

Key components of HiJiffy’s AI framework

The diagram above illustrates the key components of our artificial intelligence called Aplysia. Let’s take a look at each of them, as well as the areas where they intersect.

1. Chatbot NLP (Natural Language Processing)

• What is it?
  This component leverages Natural Language Processing to enhance the chatbot’s capability of responding to user inquiries in a more human-like way.

• What’s its role in the framework?
  It ensures the chatbot can interpret guest inquiries accurately and provide smooth, clear interactions, enhancing the overall experience.

2. Advanced AI capabilities

• What are they?
  This component combines complex AI technologies, including generative AI for formulating responses, AI text refinement to assist Console agents, anti-hallucination mechanisms for reliability, and the AI Knowledge Scanner for document-based learning.

• What is their role in the framework?
  Advanced AI capabilities improve the chatbot’s accuracy and adaptability, help Console agents deliver high-quality service, and ensure the system is robust and reliable.

3. Data analysis

• What is it?
  This component focuses on exploring data that resulted from (prospective) guest interactions and other sources. 

• What’s its role in the framework?
  Data analysis, such as Reports in the Console, is an essential and powerful tool for hoteliers as it provides actionable insights and can identify trends. Hotel teams can use them to better understand guest behaviour and preferences, leading to more informed decision-making and personalised guest experiences.

Intersection areas: powerful synergies at your disposal

• Chatbot NLP + Advanced AI:
  Enhancing the chatbot’s ability to generate accurate, contextually relevant responses and refine agent interactions.

• Advanced AI + Data analysis:
  Predicting trends, identifying patterns, and generating actionable insights, improving strategy and decision-making processes.

• Data analysis + Chatbot NLP:
  Informing the chatbot’s responses with contextually relevant insights derived from guest data, ensuring personalised and effective interactions.

Central overlap: where it all comes together

• The place in the diagram where all circles overlap represents where your Guest Communications Hub is. That’s where our AI framework brings all functionalities to work together, enhancing capabilities.

Key components breakdown: Powerful tools at your fingertips

HiJiffy’s solution enhances the guest experience and automates hotel operations. Each key component we explored above powers a range of features and capabilities. Let’s have a look at some of the most popular ones among the users of HiJiffy:

1. Chatbot functionality and NLP

Chatbot AI Features:

• Message categorisation models: Labelling incoming messages in the in-stay solution into three main types: requests, complaints, and general questions.

• Smart property identification: Detecting specific hotels or properties based on conversations with (prospective) guests for streamlined bookings and inquiries.

• Sentence summarisation: Summarising long sentences for clearer communication.

• Translations: Supporting 130+ languages enhances its ability to operate effectively in a global market.

Natural Language Processing (NLP):

• Proprietary NLP model: Developed based on six years of hospitality-specific data, to achieve the highest level of accuracy.

• Self-learning mechanism: Integrating the Console users’ feedback to enhance the performance of the NLP model.

• Complementary tools: Other machine learning and deep learning systems that complement our proprietary NLP model.

Voice assistant:

• AI-powered voice assistants: Enabling guests to communicate with hotels via voice notes and voice messages.

2. Advanced AI capabilities

Generative AI:

• Enhanced responses: Generating appropriate and, more importantly, correct responses based on the context.

• AI text refinement: Tools to assist the agents in providing the best customer service.

• Anti-hallucination mechanisms: Algorithms and models to avoid generative AI to hallucinate.

AI Knowledge Scanner:

• Document upload: By uploading a knowledge base document, the chatbot can answer common, frequently asked questions (FAQs) about the hotel and its services as fast and relevantly as possible.

3. Data analysis and behavioural insights

• Data analytics: Displaying all key metrics to hotel managers across our entire platform.

Behavioural analysis:

• Tone classification: Identifying message types to understand the context and tone of guest messages in social media conversations.

• Insights feedback: Collecting feedback to improve chatbot response quality and guest satisfaction.

Sentiment Analysis:

• Automatic labelling: Marking conversations as positive, negative, or neutral to prioritise and address urgent queries.

Learn more about our AI

If you want to learn more about other technologies used in our proprietary AI, Aplysia, explore a section of our website dedicated to our artificial intelligence, follow HiJiffy on LinkedIn and subscribe to our newsletter in the footer.

Sources

This article is based on technical contributions by Eduardo Machado from HiJiffy’s AI Team.