This week I completed writing my thesis and I’m waiting for feedback

This week I started evaluating the data I collected. I grouped the open-ended questions in one table and used NVIVO for the thematical analysis. It was my first time using Nvivo, with a bit of research, I learned how to upload the data and create different codes. With practice, I understood the different themes and…

For the past week, I have been conducting my user testing. Students from my class and from the CCI cohort took part, as well as student from LCC and people outside of university. I collected 46 responses for the AI interaction survey and 44 for the AI embodiment survey. I was expecting to have the…

The 2 following validated questionnaire will be integrated in my survey. 1. Pleasure, Arousal, Dominance (PAD): https://pdodds.w3.uvm.edu/research/papers/others/2014/bakker2014a.pdf https://thejsms.org/index.php/JSMS/article/view/230/118 Pleasure, arousal, and dominance are considered three fundamental dimensions of emotional response that capture the full range of human feelings. In my project, these dimensions are essential for evaluating how different AI personalities and emotional expressions influence…

Based on the beneficiaries section and literature review these are the points needed to be addressed in the questionnaire to meet the study’s goals: I conducted initial research on different validated questionnaires I could use in my study. The questionnaires below are from https://arxiv.org/html/2403.00582v1 : 1. Trust Scale for the AI Context (TAI): This questionnaire…

Research Initially, I considered both React.js and Streamlit for this experiment. Feature Streamlit React.js Ease of Use Simple, Python-based, great for quick prototypes Requires more setup and knowledge of JavaScript, but highly flexible Development Speed Fast, with built-in components and automatic UI generation Slower initial setup, but highly customisable once set up Customisability Limited; mostly…

In this study, the focus is not on the gender of the avatar but rather on the style of its representation. The avatars explored fall into four distinct categories based on their visual and behavioural characteristics: To create videos of the talking avatars from static images, I used three different online tools, each selected for…

Testing and iterations 1. Understanding and replicating emotions and personality traits After coming up with multiple ways for the AI tool to understand the user’s personality and emotions, I found that analysing linguistic features was the most accurate and efficient approach. I tried prompting different LLMs to understand the user’s emotions and personality. To start,…

Important points from week 3: Questions for the experiment: In the experiment, I want the user to have a seamless conversation with the LLM, as they would in their daily life with ChatGPT. However, I would like to break the illusion that the LLM understands their emotions and show its true colours. Show that it…

This week, I was researching how to implement personality features and emotion detection in LLMs. 1. I looked into a simple emotion detection from sentences. https://colab.research.google.com/drive/1Xq9u33gEpeENiPCM-hyI-DeQEq6SblWx https://github.com/Codewello/Emotion-Detection/blob/main/RoBERTa/Emotions_RoBerta.ipynb https://youtu.be/qd7IBDvgxwQ?si=NZ2Jxz4sA0T9uMhG The example below shows the score of different emotions associated with the input sentence. classifier(“This is very awful to watch!”) [[{‘label’: ‘anger’, ‘score’: 0.010571349412202835}, {‘label’: ‘disgust’, ‘score’: 0.9500361084938049}, {‘label’: ‘fear’, ‘score’: 0.029230549931526184},…

How can an AI agent customise interactions based on the individual’s traits (personality, emotions) to influence their behaviour and drive desired outcomes? Goal of the output: Show how AI’s human-like persona influences user engagement and behaviour, aiming to raise awareness of the ethical implications of anthropomorphic AI design and its impact on perceptions and user…

I focused my research on human-AI interaction. These are the first papers I read and the main insights: This helped me recognise the importance of designing ethical AI systems that users trust. They also sparked my curiosity about how AI embodiment influences user perception and interaction, leading me to explore my next research topic: anthropomorphism…

My initial research focused on gesture and sensor-based interactions, exploring how these technologies can enhance human-machine interaction (HMI) systems. I researched ways to create more intuitive and responsive interfaces, enabling seamless communication between users and products. I investigated various use cases, from home automation to autonomous vehicles, as well as context-aware computing to enhance personalisation…