Journaling is a widely adopted technique, known to improve mental health and well-being by enabling reflection on past events. Large amounts of text in digital journaling applications could hinder the reflection process due to information overload. Abstractive summarization can solve this problem by generating short summaries to quickly glance at and reminisce. In this paper, we present an investigation of the utility of large language models in the context of autobiographical text summarization. We study two approaches to adapt a self-supervised learning (SSL) model to the domain of autobiographical text. One model employs transfer learning using our new autobiographical text summary dataset to fine-tune the SSL model. The second model leverages existing news datasets for high-quality text summarization mixed with our autobiographical summary dataset. We conducted mixed methods research to analyze the performance of these two models. Through objective evaluation using ROUGE and BART scores, we find that both these approaches perform significantly better than the SSL model fine-tuned with only high-quality news datasets, showing the importance of domain adaptation and autobiographical text summary dataset for this task. Secondly, through a subjective evaluation on a crowd-sourcing platform, we evaluated the summaries generated from these models on various quality criteria such as grammar, non-redundancy, structure, and coherence. We found that on all criteria, these summaries score >4 out of 5, and the two models show comparable results. We deployed a proof-of-concept web-based journaling application to assess the practical real-world implications of incorporating abstractive summarization in a digital journaling context. We found that the participants showed a high consensus that the summaries generated by the system captured the main idea of their journal entry (80% of the 75 participants gave a Likert scale rating of ≥5.0 out of 7.0, with the overall mean rating of 5.56 ± 1.32) while being factually correct, and they found it to be a useful feature of a journaling application. Finally, we conducted human evaluation studies to compare the quality of the summaries generated from a commercial tool ChatGPT and mixed distribution fine-tuned SSL model, and present insights into these systems in the context of autobiographical abstractive text summarization. We have made our model, dataset, and subjective evaluation questionnaire openly available to the research community.

This paper describes the experimental framework and results of the IJCB 2022 Mobile Behavioral Biometrics Competition (MobileB2C). The aim of MobileB2C is bench-marking mobile user authentication systems based on behavioral biometric traits transparently acquired by mobile devices during ordinary Human-Computer Interaction (HCI), using a novel public database, BehavePassDB11https://github.com/BiDAlab/MobileB2C_BehavePassDE, and a standard experimental protocol. The competition is divided into four tasks corresponding to typical user activities: keystroke, text reading, gallery swiping, and tapping. The data are composed of touchscreen data and several background sensor data simultaneously acquired. “Random” (different users with different devices) and “skilled” (different user on the same device attempting to imitate the legitimate one) impostor scenarios are considered. The results achieved by the participants show the feasibility of user authentication through behavioral biometrics, although this proves to be a non-trivial challenge. MobileB2C will be established as an on-going competition22https://sites.google.com/view/mobileb2c/.

Emotional Self-Awareness (ESA) plays a vital role in physical and mental well-being. Recent advancements in artificial intelligence technologies have shown promising emotion recognition results, opening new opportunities to build systems to support ESA. However, little research has been done to understand users' perspectives on artificial-intelligence-based emotion recognition systems. We introduce Troi, an automatic emotion recognition mobile app using wearable signals. With Troi, we ran a multi-day user study with 12 users to understand user preference parameters, such as perceived accuracy, confidence, preferred emotion representations, effect of self-awareness of emotions, and real-time use cases. Further, we extend our study to evaluate the machine learning model in-the-wild to understand behaviours in-the-wild. We found that users perceived accuracy of the emotion recognition model is higher than the actual model prediction accuracy; there was no strong preference for one specific emotion representation, and users' self-awareness of emotions improved over time.

Self-Tracking stress at work is an important aspect of stress management and is often the first step to improving mental health and personal well-being. We introduce StressShoe, a DIY toolkit for self-tracking stress and just-in-time stress interventions for office workers performing sedentary tasks. Informed by a focus group study and a pilot study, we designed and tested a toolkit that consists of a single, off-the-shelf, shoe-mounted inertial measurement unit (IMU), a machine learning model that estimates acute stress, and a companion mobile app. The mobile app allows for user-defined just-in-time interventions on the estimated stress level. To demonstrate the benefits of our system we evaluated StressShoe with 10 users over 4 weeks. We identified several effective just-in-time personalised intervention users created and found that StressShoe is easy to use and helped them reflect on their daily stressful experiences.

Speech Emotion Recognition (SER) has been a challenging task on which researchers have been working for decades. Recently, Deep Learning (DL) based approaches have been shown to perform well in SER tasks; however, it has been noticed that their superior performance is limited to the distribution of the data used to train the model. In this paper, we present an analysis of using autoencoders to improve the generalisability of DL based SER solutions. We train a sparse autoencoder using a large speech corpus extracted from social media. Later, the trained encoder part of the autoencoder is reused as the input to a long short-term memory (LSTM) network, and the encoder-LSTM modal is re-trained on an aggregation of five commonly used speech emotion corpora. Our evaluation uses an unseen corpus in the training & validation stages to simulate 'in the wild' condition and analyse the generalisability of our solution. A performance comparison is carried out between the encoder based model and a model trained without an encoder. Our results show that the autoencoder based model improves the unweighted accuracy of the unseen corpus by 8%, indicating autoencoder based pre-training can improve the generalisability of DL based SER solutions.

Currently, smartwatches are equipped with Photoplethysmography (PPG) sensors to measure Heart Rate (HR) and Heart Rate Variability (HRV). However, PPG sensors consume considerably high energy, making it impractical to monitor HR & HRV continuously for an extended period. Utilising low power accelerometers to estimate HR has been broadly discussed in previous decades. Inspired by prior work, we introduce CompRate, an alternative method to measure HR continuously for an extended period in low-intensity physical activities. CompRate model calibrated for individual users only has an average performance of Root Mean Squared Error (RMSE) 1.58 Beats Per Minute (BPM). Further, CompRate used 3.75 times less energy compared to the built-in PPG sensor. We also demonstrate that CompRate model can be extended to predict HRV. We will demonstrate CompRate in several application scenarios: self-awareness of fatigue and just-in-time interruption while driving; enabling teachers to be aware of students’ mental effort during a learning activity; and the broadcasting of the location of live victims in a disaster situation.