Personal and Ubiquitous Computing最新文献

Free-text keystroke dynamics, the unique typing patterns of an individual, have been applied for the security of mobile devices by providing the non-intrusive and continuous user authentication. Existing authentication approaches mainly concentrate on the keystroke dynamics when operating a specific device, and overlook the generality of keystroke dynamics for cross-device user authentication. To tackle this problem, in this paper, we propose an efficient federated free-text keystroke dynamics mechanism to mitigate the difference in keyboards for cross-device authentication. Specifically, we explore and analyze the keystroke features of various keyboards and extract cross-device keystroke features. To protect user privacy, their type of rhythm information must be kept locally. We utilize federated learning based on the auxiliary model to train the authentication model. Our proposed solution was evaluated on a large-scale data set with 168,000 users. The experimental results show that our proposed solution performs well with great robustness across different types of keyboards.

In the growing field of sonic interaction design, increasing emphasis is being placed on walking-based interactions within an array of applications, including virtual reality, interactive media, and rehabilitation. Our study focuses on recreating the aural experience of wading through water, specifically the challenge of eliciting accurate and natural movement-sound associations for wading, both in interactive and listening-only contexts. We engineered a real-time wading simulation using a digital Foley technique that maps lower limb angular velocity contours to the intensity of steady-state flowing sounds. Our first iteration was evaluated in a perceptual experiment involving 16 participants, as well as an interactive test with 9 participants, which revealed the need for additional sensors along with time-domain preprocessing to ensure a consistently natural sound envelope across walking cadences (step tempi). We then refined the mapping function and incorporated more sound layers. In our subsequent listening test, 55 participants compared the realism of the initial and refined versions with real-life wading sounds at various step cadences. While the refined version demonstrated a notable improvement over the initial one and was deemed fairly realistic overall, it fell just short of the authenticity of the real-life recordings at faster cadences, suggesting room for further improvement of our simulation. Nevertheless, this research marks a stride in the evolution of walking-based sonic interactions, instigating wider acceptance and application of such systems in the future.

Conversational technologies have become increasingly prevalent, with interactions becoming more and more personalized. This paper reviews literature on the phenomenon of self-disclosure to conversational technologies. Five types of factors emerge as influencing self-disclosure: interface modality, conversational factors, user characteristics, mediating mechanisms, and contextual factors. We describe each type of factor, cover findings from the literature, present the framework of factors influencing self-disclosure that thus emerges, and put forth pertinent questions for future research on self-disclosure to conversational technologies.

Companies increasingly rely on chatbots to enable efficient and engaging communication with customers. Previous research has highlighted a trend towards female-gendered designs of customer service chatbots, adding to concerns about the reinforcement of outdated gender stereotypes in human-computer interactions. Against this background, the present study explores design alternatives to traditionally gendered chatbot avatars. In an online experiment, N = 1064 participants interacted with a bank service chatbot, where one half saw a gender-ambiguous anthropomorphic face as the chatbot’s default avatar, and the other half an abstract non-anthropomorphic icon. Contrary to earlier studies, which linked anthropomorphism to higher user acceptance, our manipulation of avatars did not significantly alter intentions to use the chatbot. After the interaction, participants could select their preferred avatar image from a set of six, including non-anthropomorphic icons (speech bubbles) and anthropomorphic faces (female, male, gender-ambiguous). While many adhered to their initially viewed image, a clear majority opted for abstract non-anthropomorphic icons. This overall preference was consistent across all user genders, although men were more likely than women to favor a traditionally female-looking avatar. Notably, less than a quarter of participants recognized the gender-ambiguous avatar as such. In accordance with traditional gender binaries, most identified it as either male or female. Those who perceived it as female reported higher intentions to use the chatbot. As a practical implication, our findings advocate for the adoption of more abstract and gender-neutral chatbot designs, as they not only help to avoid problematic stereotypes but also seem to align with customer preferences for non-gendered chatbot interactions.

Reflecting on everyday experiences offers valuable insights and has the potential to enhance psychological well-being. Yet, only some have access to a facilitator for reflection. Conversational agents hold promise as companions for these discussions. We surveyed individuals with therapy experience to understand user needs and arrived at interaction strategies used in therapy. We then evaluated these strategies with five therapists and transformed our data, along with their input, into a set of interaction strategies to be used on conversational agents for reflection. We developed an AI chatbot prototype where we implemented these strategies and conducted a 1-week in-the-wild study with 34 participants to evaluate the interaction strategies and experiences of interacting with a chatbot for reflection. Findings reveal that participants are willing to engage with a chatbot, even with limited capabilities. Critical aspects include the chatbot’s contextual awareness, statement repetition, and human-like qualities. Successfully balancing questions with non-question statements is essential for a pleasurable dialogue-driven reflection. Our paper presents implications for future design and research studies.

Smart speakers are entering our homes and enriching the connected ecosystem already present in them. Home inhabitants can use those to execute relatively simple commands, e.g., turning a lamp on. Their capabilities to interpret more complex and ambiguous commands (e.g., make this room warmer) are limited, if not absent. Large language models (LLMs) can offer creative and viable solutions to enable a practical and user-acceptable interpretation of such ambiguous commands. This paper introduces an interactive disambiguation approach that integrates visual and textual cues with natural language commands. After contextualizing the approach with a use case, we test it in an experiment where users are prompted to select the appropriate cue (an image or a textual description) to clarify ambiguous commands, thereby refining the accuracy of the system’s interpretations. Outcomes from the study indicate that the disambiguation system produces responses well-aligned with user intentions, and that participants found the textual descriptions slightly more effective. Finally, interviews reveal heightened satisfaction with the smart-home system when engaging with the proposed disambiguation approach.

This study developed a Human-Centered Technology Acceptance Model (HC-TAM) for recruitment chatbots, integrating aspects of the traditional Technology Acceptance Model (TAM)(Davis in 1989) with a focus on human-centered factors such as transparency, personalization, efficiency, and ethical concerns, alongside the fundamental TAM constructs of perceived ease of use and perceived usefulness. The study shows that the intention to use technology is influenced by their perceptions of its usefulness and ease of use. By extending TAM to include human-centered considerations, this research aimed to capture the diverse factors that significantly influence users’ acceptance of chatbots in the recruitment process. A three-phase study has been carried out, each serving a distinct purpose. (a) Phase 1 focuses on defining primary themes through qualitative interviews with 10 participants, laying the foundation for subsequent research. (b)Building upon this foundation, Phase 2 engages 28 participants in a refined exploration of these themes, ending in a comprehensive landscape of user perspectives. (c) Finally, Phase 3 employs rigorous Structural Equation Modeling for theoretical framework examination, yielding critical constructs and hypotheses. Moreover, Phase 3 encompasses the thorough development of measurement instruments and extensive data collection, involving 146 participants through questionnaires, the study found that the acceptance of recruitment chatbots is significantly enhanced when these systems are designed to be transparent, provide personalized interactions, efficiently fulfill user needs, and address ethical concerns. These findings contribute to the broader understanding of technology acceptance in the context of recruitment, offering valuable insights for developers and designers to create chatbots that are not only technically advanced but also ethically sound, user-friendly, and effectively aligned with human needs and expectations in recruitment settings.

The proliferation of sensors and smart Internet of Things (IoT) devices in our everyday environments is reshaping our interactions with everyday objects. This change underlines the need to empower non-expert users to easily configure the behaviour of these devices to align with their preferences and habits. At the same time, recent advances in generative transformers, such as ChatGPT, have opened up new possibilities in a variety of natural language processing tasks, enhancing reasoning capabilities and conversational interactions. This paper presents RuleBot + + , a conversational agent that exploits GPT-4 to assist the user in the creation and modification of trigger-action automations through natural language. After an introduction to motivations and related work, we present the design and implementation of RuleBot + + and report the results of the user test in which users interacted with our solution and Home Assistant, one of the most used open-source tools for managing smart environments.

We examined how politeness perception can change when used by a human or voice assistant in different contexts. We conducted two norming studies and two experiments. In the norming studies, we assessed the levels of positive politeness (cooperation) and negative politeness (respecting autonomy) conveyed by a range of politeness strategies across task (Norming Study 1) and social (Norming Study 2) request types. In the experiments, we tested the effect of request type and imposition level on the perception of written requests (Experiment 1) and requests spoken by a voice assistant (Experiment 2). We found that the perception of politeness strategies varied by request type. Positive politeness strategies were rated as very polite with task requests. In contrast, both positive and negative politeness strategies were rated as very polite with social requests. We also found that people expect agents to respect their autonomy more than they expect them to cooperate. Detailed studies of how request context interacts with politeness strategies to affect politeness perception have not previously been reported. Technology designers might find Tables 4 and 5 in this report especially useful for determining what politeness strategies are most appropriate for a given situation as well as what politeness strategies will evoke the desired feeling (autonomy or cooperation).

In today’s business landscape, Chatbots play a pivotal role in innovation and process optimization. In this paper, we introduced a novel advanced Emotional Chatbot AI, introducing sentiment analysis for human chatbot conversations. Adding an emotional component within the human-computer interaction, can in fact dramatically improve the quality of the final conversation between Chatbots and humans. More specifically, in our paper, we provided a practical evaluation of the EmoROBERTA software, introducing it into a novel implementation of an Emotional Chatbot. The pipeline we present is novel, and we developed it within a business context in which the use of sentimental and emotional responses can act in a significant and fundamental way toward the final success and use of the Chatbot itself. The architecture enriches user experience with real-time updates on the topic of interest, maintaining a user-centric design, toward an affective-response enhancement of the interaction established between the Chatbot and the user. The source code is fully available on GitHub: https://github.com/filippoflorindi/F-One.