Intelligent User Interfaces Flashcards
Intelligent User Interfaces: In the UI
- In the UI: Intelligent processing is found in the user interface(s) of the system, and its purpose is to enable effective, natural, or otherwise appropriate interactions with users of the system
- The user interfaces are automatically adapted to the inferred capabilities, needs of the user or context of the interaction (e.g. Gmail Smart Compose)
- Multimodal and sensing-based systems that aim to enable more natural, human-like forms of input and output (e.g. Speech, Body language)
Intelligent User Interfaces: Behind the UI
- Behind the UI: Intelligent processing is found in the backend of the system, and its purpose is to serve some beneficial functions, such as performing actions on behalf of the user or modelling the user for personalisation
- Recommender systems that provide us with personalised options (e.g. Netflix, Spotify)
- Agents that perform complex or repetitive tasks with some guidance from the user
- Situated assistance systems that monitor and support user activities (e.g. Health andactivity monitoring)
Intelligent User Interfaces
- IUIs employ a variety of AI technologies in order to assist users achieve their goals and
requirements - IUIs assist users by …
- using AI to optimise the interaction by adapting to the user and their context
- using AI to enable users to interact with the interface similar to how humans interact, making it more intuitive and user-friendly
- using AI to intelligently provide personalised and relevant suggestions, recommendations, and responses based on their individual preferences and needs
Example: Precise Touch
- How accurate is touch input our smartphones?
- Touch is inherently inaccurate on smartphone screens
- Adapt keyboard to typist
- Probilistic key model
- Language model influence (e.g. bigram for english)
- Context adaptations
-Using AI to improve touch input itself
Combining AI to optimise interaction
- Combination of AI technologies employed for precise touch
- User modelling to adapting the keyboard based on the typing behaviour of the user
- Machine learning and data from other sensors to determine the context of the user
- Computer vision and deep learning to increase the precision of the touch input
- Natural language processing, where a language model is to predict the likelihood of what the user will type based on the language and the key that they are likely to hit next
- IUIs are often designed to be invisible to the user
- i.e. integrated seamlessly into the user experience, allowing users to interact with the
technology in a natural and intuitive way without being aware of the underlying technology - Combination of using AI in and behind the UI to create such experiences
- i.e. integrated seamlessly into the user experience, allowing users to interact with the
Smart Compose and Replies
Natural Language Processing
- Natural Language Processing (NLP) is a field of artificial intelligence (AI) that focuses on the ability of computers to understand and interpret human language, and to read and interpret text to extract information
- Have you used NLP today?
- Every time you search for something and suggestions appear (autocomplete)
- Every time you make a mistake while typing (autocorrect)
- Every time you use a virtual assistant (speech-to-text, text understanding)
- Key enabling technology for many IUIs
- Examples: virtual assistants (Siri, Alexa), chatbots, search engines
Natural Language Processing - Timeline
Natural Language Processing - Evolution
NLP: Large Language Models
- Large language models can generate human-like text, trained on amounts of text data in order to learn the patterns and structures
of human language - Current large language models cannot tell fact from fiction yet, resulting in biased and incorrect results, or even false and misleading information
- Space bears, it’s easy to spot fiction when it involves space bears, but harder with a subject users may not know much about
- Can you trust an IUI built on a large language model that may provide you with false or misleading information?
Natural Language Processing
- Other applications areas of natural language processing in IUI context
- AI writing tools - Copyai (text generation), Grammarly (text editor), Quillbot (text summarisation)
- Chatbots and virtual personal assistants - Siri, Alexa, Cortana, Messenger
- Search engines - Google, Bing
- Automated language translators -
Google Translate - Social media - Twitter (text analysis for
personalised content)
Recommender Systems
- The goal of a recommender system is to generate meaningful recommendations to a collection of users for items or products that might interest them
- Why are recommender systems used?
- The system has more information that it can present (millions of options to display)
- Collecting a subset of items out of a bigger number of items (collection of songs)
- Ordering items from a subset - Which one should be shown first?
- How do recommender systems impact the user experience?
- If it’s perfect: This is “exactly” what I wanted.
Recommender Systems #2
- User modelling refers to the process of collecting and analysing data about a user’s abilities, preferences, and behavior in order to create a model of their characteristics and needs
- Collaborative filtering is a technique that uses the ratings and preferences of all users in a
system to make recommendations- “people who liked what you liked also liked X, hence I suggest X to you”
- Content-based filtering is a technique that uses the characteristics of the items in a system
(such as the genres, actors, and directors of movies in a streaming video service) to recommend similar items to a user.- “X is similar (based on similar measure for a domain) to what you liked before, hence I
suggest X to you”
- “X is similar (based on similar measure for a domain) to what you liked before, hence I
What are the benefits of using IUI
- Provide personalised experiences
- Prevent information overload
- Provide proactive assistance
- Enables natural interaction
Benefit: Provide personalised experiences
No two persons are the same and people have different habits, preferences, and working methods and environments. IUIs that take these differences into account can provide a personalised method of interaction. The interface knows the user and can use that knowledge in its communication with the user.