The early prediction of depressive mood episodes is crucial for effective intervention in patients with Major Depressive Disorder (MDD) and Bipolar Disorder (BD). This study explores a predictive framework leveraging digital phenotypic data collected from smartphones and smartwatches, with a focus on circadian rhythm indicators such as Dim Light Melatonin Onset (DLMO). Using data from 164 participants within the Mood Disorder Cohort Research Consortium in Korea, time-series features related to sleep, heart rate, activity levels, and light exposure were processed to predict mood episodes seven days in advance. Deep learning models, including LSTM, GRU, and an LSTM-GRU hybrid, were applied to analyze this data, with the GRU model achieving the highest recall (0.767) and the LSTM model displaying superior robustness across metrics. SHAP value analysis of DLMO-related variables further underscored the association between circadian rhythm disruptions and depressive episodes, with delayed wake-up times relative to ideal schedules linked to increased depressive symptoms. Our findings demonstrate the feasibility of using digital phenotypes for early detection of mood episodes. These results highlight the potential of automated monitoring systems in clinical practice, which enable proactive intervention strategies through continuous, objective monitoring of patient conditions.

To facilitate software quality assurance, defect prediction metrics, such as source code metrics, change churns, and the number of previous defects, have been actively studied. Despite the common understanding that developer behavioral interaction patterns can affect software quality, these widely used defect prediction metrics do not consider developer behavior. We therefore propose micro interaction metrics (MIMs), which are metrics that leverage developer interaction information. The developer interactions, such as file editing and browsing events in task sessions, are captured and stored as information by Mylyn, an Eclipse plug-in. Our experimental evaluation demonstrates that MIMs significantly improve overall defect prediction accuracy when combined with existing software measures, perform well in a cost-effective manner, and provide intuitive feedback that enables developers to recognize their own inefficient behaviors during software development.

ABSTRACT Botnets are a disastrous threat because they execute malicious activities such as distributed denial‐of‐service, spam email, malware downloads (such as eggdownloads), and spying by exploiting zombie PCs under their control. Botnets infect PCs on a huge scale by initially scanning the service ports of vulnerable applications for the purpose of propagation, which is leveraged as the size of the botnet increases. Therefore, it is of crucial importance to detect botnet‐propagation activities early and to determine the expectedsize of the attack. To address this issue, this paper proposes to recreate botnets' port‐scanning patterns using a simple text classifier that represents these patterns as a kind of matrix. The patterns obtained are then used to train a hidden Markov model and to perform early detection using the trained model. Early detection is achievable by catching the onset of suspicious propagation immediately, and a size estimate is obtained by monitoring fluctuations in botnet size. With this approach, early‐detection rates increased to more than 30.6% on average, with a low false negative rate (less than 6%) and an F‐measure greater than 96%. This significant improvement in performance will contribute to preventing botnet propagation in its earliest stages. Copyright © 2011 John Wiley & Sons, Ltd.

There is a common belief that developers' behavioral interaction patterns may affect software quality. However, widely used defect prediction metrics such as source code metrics, change churns, and the number of previous defects do not capture developers' direct interactions. We propose 56 novel micro interaction metrics (MIMs) that leverage developers' interaction information stored in the Mylyn data. Mylyn is an Eclipse plug-in, which captures developers' interactions such as file editing and selection events with time spent. To evaluate the performance of MIMs in defect prediction, we build defect prediction (classification and regression) models using MIMs, traditional metrics, and their combinations. Our experimental results show that MIMs significantly improve defect classification and regression accuracy.

Battery exhaustion attack is a ldquodenial of servicerdquo attack on mobile devices, causing more battery discharge than normally necessary. The excessive discharge arises out of the limitation in battery capacity. Researches have been conducted to discover the cause of the abnormal usage of handset battery by employing power estimation techniques. It is also important, however, to automatically respond to the attack and maintain the power discharge at the normal level. In this study, we introduce a method designed to detect the defective process and to self-correct the abnormality by stopping or delaying the suspicious actions. The proposed system monitors system calls and the power usage of a local process, and delays activation of the abnormal system calls until the system calls can be run without hampering the battery life. The benefit of our proposed method lies in that it autonomously protects the expected battery lifespan during a battery exhaustion attack without shutting down operation of falsely detected programs.

Over the past few years, a considerable number of studies have been made on the Return on Security Investment (ROSI) in connection with security policy. However, it is necessary to quantity types of projected damages, cyber threats, security safeguards and ROSI, since the size of information system and its investment factors are growing continuously. Accordingly, in this paper, we propose a way of effective security investment for a higher ROSI, focusing on how to efficiently conduct assessment with use of the Analytic Hierarchy Process (AHP)-based Canonical Correlation Analysis (CCA). The CCA is a means of measuring the linear relationship between relevant cyber threats and corresponding security-safeguarding methods. Finally, we illustrate how such analyses can be used for determination of which security-safeguarding method functions most efficiently, as alternative to the current security investment policy.

Prevention is normally recognized as one of the best defense strategy against malicious hackers or attackers. The desire of deploying better prevention mechanisms has motivated many security researchers and practitioners, who are studies threat trend analysis models. However, threat trend is not directly revealed from the time-series data because the trend is implicit in its nature. Besides, traditional time-series analysis, which predicts the future trend pattern by relying exclusively on the past trend pattern, is not appropriate for predicting a trend pattern in dynamic network environments (e.g., the Internet). Thus, supplemental environmental information is required to uncover a trend pattern from the implicit (or hidden) raw data. In this paper, we propose cyber threat trend analysis model using hidden Markov model (HMM) by incorporating the supplemental environmental information into the trend analysis.