Kazjon Grace

Investigating intuitiveness and effectiveness of gestures for free spatial interaction with large displays

A key challenge for creating large interactive displays in public spaces is in the definition of ways for the user to interact that are effective and easy to learn. This paper presents the outcomes of user evaluation sessions designed to test a series of different gestures for people interacting with large displays in the public space. It is an initial step towards the broader goal of establishing a natural means for immersive interactions. The paper proposes a set of simple gestures for the execution of the basic actions of selecting and rearranging items in a large-scale dashboard. We performed a comparative analysis of the gestures, leading to a more in-depth understanding of the nature of spatial interaction between people and large public displays. More specifically, the analysis focuses on the scenarios when the interaction is restricted to an individuals own body, without any further assistance from associated devices. The findings converge into the elaboration of a model for assisting with the applicability of spatial gestures in response to both the context and the content they are applied to.

Modeling Expectation for Evaluating Surprise in Design Creativity

This paper describes methods for characterizing expectation as it applies to calculating surprise for evaluating the creativity of designs. Building on a model of creative designs as being novel, surprising and valuable we develop a typology of the kinds of expectations that, when violated, produce surprise. Contrasting computational models of two of kinds of expectation are presented in the domain of mobile devices and their respective advantages for creativity evaluation are described.

Data-intensive evaluation of design creativity using novelty, value, and surprise

The increasing availability of large quantities of product-related data provides an opportunity to augment human designers with analytical models for evaluating the creativity of a new design. In this paper we describe three characteristics of a creative design: novelty, value, and surprise. We present an analytical framework for computationally evaluating the creativity of a new design. Building on our previous work, we associate each characteristic with a computational process, and develop a new model for evaluating surprise using predictive analytics. We describe an implementation of our analytical models as applied to a data-set of mobile devices. We report on the most surprising devices identified by our models and their corresponding novelty and value scores, and conclude by discussing the broader applications and implications of an analytical approach to evaluating creative designs.

Combining CBR and Deep Learning to Generate Surprising Recipe Designs

This paper presents a dual-cycle CBR model in the domain of recipe generation. The model combines the strengths of deep learning and similarity-based retrieval to generate recipes that are novel and valuable (i.e. they are creative). The first cycle generates abstract descriptions which we call “design concepts” by synthesizing expectations from the entire case base, while the second cycle uses those concepts to retrieve and adapt objects. We define these conceptual object representations as an abstraction over complete cases on which expectations can be formed, allowing objects to be evaluated for surprisingness (the peak level of unexpectedness in the object, given the case base) and plausibility (the overall similarity of the object to those in the case base). The paper presents a prototype implementation of the model, and demonstrates its ability to generate objects that are simultaneously plausible and surprising, in addition to fitting a user query. This prototype is then compared to a traditional single-cycle CBR system.

A Process Model for Crowdsourcing Design: A Case Study in Citizen Science

Crowdsourcing design has been applied in various areas of graphic design, software design, and product design. This paper draws on those experiences and research in diversity, creativity and motivation to present a process model for crowdsourcing experience design. Crowdsourcing experience design for volunteer online communities serves two purposes: to increase the motivation of participants by making them stakeholders in the success of the project, and to increase the creativity of the design by increasing the diversity of expertise beyond experts in experience design. Our process model for crowdsourcing design extends the meta-design architecture, where for online communities is designed to be iteratively re-designed by its users. We describe how our model has been deployed and adapted to a citizen science project where nature preserve visitors can participate in the design of a system called NatureNet. The major contribution of this paper is a model for crowdsourcing experience design and a case study of how we have deployed it for the design and development of NatureNet.

Gesture-based interaction design: communication and cognition

This workshop explores and identifies the cognitive issues fundamental to the design of gestural interactive systems. To achieve this, a dialogue will be facilitated among researchers in the cognitive science of gesture and gestural interaction within the HCI community. During the workshop we will discuss the different methodologies and results within the study of gestural interaction, with a focus on how the use of bodily movement in an interface affects the cognition of users, groups, communities and societies. We invite participants from cognitive science, HCI, user experience design, educational technology and interactive installation art to present their work on gestural interfaces and discuss how that work has been observed to impact user perceptual or cognitive faculties. The workshops material outcomes include a book on gestural interaction and cognition, while the research outcomes include methodologies, heuristics, design principles and hypotheses for the further design and investigation of gestural and tangible technologies.

NatureNet: a model for crowdsourcing the design of citizen science systems

NatureNet is citizen science system designed for collecting bio-diversity data in nature park settings. Park visitors are encouraged to participate in the design of the system in addition to collecting bio-diversity data. Our goal is to increase the motivation to participate in citizen science via crowdsourcing: the hypothesis is that when the crowd plays a role in the design and development of the system, they become stakeholders in the project and work to ensure its success. This paper presents a model for crowdsourcing design and citizen science data collection, and the results from early trials with users that illustrate the potential of this approach.

Personalised Specific Curiosity for Computational Design Systems

The Personalised Curiosity Engine (PQE, pronounced “pique”) is a framework for computational design systems that models the curiosity of an individual user. This model is then used to synthesise designs that stimulate that user’s curiosity. PQE extends our previous research in modelling surprise and curiosity by adding a model of a specific user to generate designs that are personally creative for that user: novel and valuable for them, but not necessarily for society. We describe PQE as a framework, and then describe Q-chef: a design system applying PQE in the domain of recipe generation with a goal of diversifying its user’s diet over time. We evaluate our framework with several simulations of Q-chef components that serve as a proof-of-concept of the role of personalised curiosity modelling in computational design.

Teaching Evolutionary Design Systems by Extending Context Free

This document reports on a case study using a novel approach to teaching generative design systems. The approach extends Context Free, a popular design grammar for producing 2D imagery, to support parametric and evolutionary design. We present some of the challenges that design students have typically faced when learning about generative systems. We describe our solution to providing students with a progressive learning experience from design grammars, through parametric design, to evolutionary design. We conclude with a discussion of the benefits of our approach and some directions for future developments.

Enticing Casual Nature Preserve Visitors into Citizen Science via Photos

While scientists need the contributions of members of the public if they are to document biological diversity across large spaces and over long periods of time, it is challenging to recruit enough volunteers. Since many people use their smartphones to take pictures when they are in nature, it may be beneficial to understand what they gravitate toward as a first step in understanding how they might be engaged in citizen science. We examined photographs taken by casual visitors to a Colorado nature preserve to look for clues about what attracts them. A thematic analysis revealed that the majority of their pictures were of plants, birds, and landscapes, and three-quarters chose to annotate some photos with comments or questions. Based on these findings, we propose ways to entice such visitors toward participating in biodiversity-oriented citizen science projects.