40 years of HCI International. Join us in Washington DC to celebrate

T06: Feeling Data Through Affective Interfaces Using Personality and Semantic Retrieval

Saturday, 29 June 2024, 13:30 - 17:30 EDT (Washington DC)
Back to Tutorials' Program

 

Jenna Fizel (short bio)

IDEO, USA

Tiange Wang (short bio)

IDEO, USA

 

Objectives:

Many current AI applications are focused on efficiency or require constant user creativity. In this tutorial, we will introduce a method that enables people to see, feel and interact with data in affective ways through a GenAI-powered interface and natural language data. We will provide a live experience around using GenAI techniques to bring new perspective to our understanding of data, add a rich emotive layer to data interpretation and interaction, and create emotionally-engaging experiences with data. The participants will gain enhanced knowledge about human-centered and emotive applications of AI, expanded view on interactive data interpretation, and hands-on experience crafting affective interfaces to natural language data.

 

Content and benefits:

In this tutorial, we will start with an overview of current AI applications, which primarily focus on efficiency, productivity, or user creativity. We will also introduce the pipeline of creating general conversational interfaces with LLMs and discuss their various characteristics. To set the stage for the subsequent section, we will review interactive data experiences that go beyond mere efficiency and utility, emphasizing the value of emotionally engaging and human-centered approaches. 
We will then describe how to use RAG (Retrieval Augmented Generation) and prompt engineering to create affective yet precise interfaces to complex data. We will ask participants to bring their own unstructured natural language data to quickly build their own conversational interface within a custom prototyping tool. Under the hood the tool uses OpenAI LLMs and functions to access custom data via Google’s Universal Sentence Encoder embeddings. We will also provide guidance and examples on how to craft contexts and conversational structures that evoke emotion, and provide examples from other work including our parking spot reservation system with a dynamic personality. 
Finally, we will discuss recent case studies and give an outlook on future developments and applications of affective data experience enabled by AI. 

 

Target Audience:

HCI professionals, Interaction and Interface Designers, Interactive Data Visualization & Data Storytelling Professionals, Developers, Researchers and Students who are interested in affective interfaces, creative data interactions, and GenAI techniques. We do not expect participants to be familiar with GenAI prototyping or programming. 

 

Additional platform or tool(s) to be used by the tutors:

ChatGPT

 

List of materials or devices required by the participants:

Computer, Natural language data (We will provide some examples but participants are encouraged to bring their own)

 

Relevant links:

https://medium.com/duct-tape-ai/whats-up-doc-f13513d20cef
https://edges.ideo.com/

 

Bio Sketch of Presenters:

Jenna Fizel is a Senior Director of Emerging Technology at IDEO. They are inspired by the unexpected ideas that arise when translating the abstract world of information into the tangible world we can see, touch and taste. At IDEO Jenna helps teams explore design questions through building with tools like AI, XR, and digital fabrication. They believe these tools can sometimes be part of the design solution—but are always useful during the design process when used as new ways of seeing. In addition to leading IDEO’s emerging technology practice, they run an 80+ person internal learning group that makes technical skills more accessible to the IDEO community, regardless of discipline or position.
They began their career in the computational geometry department of KPF, where they used their academic background at MIT to apply systems thinking to a wide range of practical problems, from value engineering construction processes to complying with building code via dynamic simulation. Somewhat surprisingly, this led them to writing their first gradient descent algorithm and falling in love with how different ways of understanding data can transform our relationship with the institutions we interact with every day. They took this love to an interactive design agency where they worked for museums, libraries and corporations designing and building mixed physical/digital spatial experiences that helped their clients express complex data to a variety of audiences. They also co-founded a fashion technology startup and served as the CTO of a technology-focused intimate apparel company.

Tiange Wang is a multidisciplinary software designer at IDEO, where she crafts human-centered and data-driven experiences spanning web/mobile platforms, XR and emerging technology, guiding clients through transformative journeys across sectors such as health, climate, media and consumer products. 
As an award-winning and internationally-exhibiting designer and creative technologist, Tiange’s work investigates innovative interactive paradigms among humans, technology and the environment. Tackling subjects like climate change, holistic wellbeing, embodied experiences and speculative futures, she employs mediums including multimodal interaction, creative data visualization, multimedia art, AR/VR, installation, food, software, AI and games.
Tiange is the recipient of UX Design Award, A’ Design Award, and Muse Creative Award among others. Her work has been exhibited and presented at Signals, Vancouver International Film Festival, SIGGRAPH, CHI Play, MOOD Museum of Design, Hebei Design Museum, Aedes Architekturforum, Boston Design Week, NYCxDesign Festival, SF Design Week, Cambridge Science Festival, and featured on Dezeen and World Design Week.
Tiange holds a Master of Architecture from Harvard Graduate School of Design and a Bachelor of Arts in Architecture with High Honors from the University of California, Berkeley.