<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:dc="http://purl.org/dc/elements/1.1/" version="2.0">
  <channel>
    <title>BURA Collection:</title>
    <link>https://bura.brunel.ac.uk/handle/2438/8631</link>
    <description />
    <pubDate>Thu, 19 Nov 2026 07:38:01 GMT</pubDate>
    <dc:date>2026-11-19T07:38:01Z</dc:date>
    <item>
      <title>Reinforcement Learning in Quantitative Trading: A Review of Current Approaches and Breakthroughs</title>
      <link>https://bura.brunel.ac.uk/handle/2438/34200</link>
      <description>Title: Reinforcement Learning in Quantitative Trading: A Review of Current Approaches and Breakthroughs
Authors: Khushi, Matloob; Sanodiya, Rakesh Kumar; Suleiman, Basem; Desa, Sachit; Uddin, Shahadat; Lu, Xiaobin
Abstract: Reinforcement learning (RL) provides an autonomous approach to quantitative trading by allowing an agent to learn a trading policy through repeated interaction with a market environment rather than through hand-crafted rules. The aim of an RL trading agent is to select actions that maximize long-term rewards while meeting a chosen level of risk. This review examines RL applications in quantitative trading, with particular attention to value-based, policy-based, and actor–critic methods. It summarizes the reported returns, risk-adjusted performance, evaluation settings, and practical limitations of the reviewed studies. Although many RL methods report promising results in backtests and simulated environments, their outcomes are highly dependent on data selection, feature construction, transaction cost assumptions, training procedures, reward design, and validation protocols. Consequently, reported returns, Sharpe ratios, and drawdowns across studies are not directly comparable and should not be interpreted as evidence that one RL method is universally superior to another. This review shows that relevant state information, suitable reward functions, realistic risk management, and deep neural feature extraction can improve RL trading performance. However, reliable progress requires transparent and consistent reporting of data handling, time-respecting validation, execution assumptions, transaction costs, slippage, and risk-adjusted trading outcomes. We therefore argue that future RL trading research should adopt Quantitative Finance Reporting Standard (QFRS), to make trading claims data leakage free, reproducible, back-tested, economically meaningful, and comparable across studies.
Description: Data Availability Statement: &#xD;
No new data were created or analyzed in this study. Data sharing is not applicable to this article.</description>
      <pubDate>Tue, 10 Nov 2026 00:00:00 GMT</pubDate>
      <guid isPermaLink="false">https://bura.brunel.ac.uk/handle/2438/34200</guid>
      <dc:date>2026-11-10T00:00:00Z</dc:date>
    </item>
    <item>
      <title>Addressing Data Scarcity and Class Imbalance Using Active Learning and Synthetic Data Generation: Experimentation on Clinical Datasets</title>
      <link>https://bura.brunel.ac.uk/handle/2438/34198</link>
      <description>Title: Addressing Data Scarcity and Class Imbalance Using Active Learning and Synthetic Data Generation: Experimentation on Clinical Datasets
Authors: Bottino, Lorella; Zucco, Chiara; Tucker, Allan; Cannataro, Mario
Abstract: The rapid advancement of Machine Learning (ML) technologies has revolutionized healthcare, particularly in the domains of disease prediction and diagnosis. Nevertheless, some data-related issues, such as annotated data scarcity, low-quality labels, and class imbalance in the dataset, can compromise the effectiveness of the learning process. This paper first gives an overview of Active Learning (AL), which is a strategy in machine learning that helps models to achieve better performance using fewer labeled examples. It then reviews various strategies for addressing dataset imbalance, in particular focusing on data-level rebalancing approaches. The paper provides a review of existing works on active learning in the presence of class imbalance and on the integration of active learning with synthetic oversampling techniques. The central core of the paper is the implementation of four different strategies for data selection from publicly available clinical datasets: Random, Random + SMOTE, Uncertainty, and Uncertainty + SMOTE. We calculated and plotted balanced accuracy and average precision by iterating over 20 different seed values, using cumulative batches of labeled data, for each combination of dataset and classifier. Moreover we also carried out a statistical test to assess the extent to which the difference between the Random and Uncertainty distributions, and that between the Uncertainty and Uncertainty + SMOTE distributions, was statistically significant. Experiments show that uncertainty-based strategy can improve predictive performance compared with random acquisition, particularly with Random Forest, although the observed benefits vary across datasets, classifiers, performance metrics, and annotation budgets. The additional contribution of SMOTE is also context dependent and does not yield statistically significant improvements over uncertainty-based acquisition alone. For the WBCD dataset, it can be seen that Uncertainty + SMOTE is comparable to, or in some cases below, the uncertainty-only strategy. Metrics such as precision, recall, specificity, F1-score, AUROC and AUPRC are evaluated and provided.
Description: Data Availability Statement: The datasets used in this article are publicly available in the UCI repository and they can be downloaded here: HFCR dataset, available at https://doi.org/10.24432/C5Z89R; WBCD dataset, available at https://doi.org/10.24432/C5DW2B. Moreover, the synthetic datasets used in this article have been generated using Synthetic Minority Over-sampling Technique (SMOTE), implemented through the imbalanced-learn Python library, available at https://imbalanced-learn.org/stable/ (accessed on 5 November 2026). The synthetic dataset presented in this study is available upon request from the corresponding author.</description>
      <pubDate>Fri, 13 Nov 2026 00:00:00 GMT</pubDate>
      <guid isPermaLink="false">https://bura.brunel.ac.uk/handle/2438/34198</guid>
      <dc:date>2026-11-13T00:00:00Z</dc:date>
    </item>
    <item>
      <title>Automated Essay Marking with Data-Efficient Augmentation and Score-Token Weighted Loss</title>
      <link>https://bura.brunel.ac.uk/handle/2438/34173</link>
      <description>Title: Automated Essay Marking with Data-Efficient Augmentation and Score-Token Weighted Loss
Authors: Shakaa, Yamama; Aburumman, Nadine; Anagnostou, Anastasia; Taylor, Simon; Li, Yongmin
Abstract: Automated Essay Scoring (AES) research has predominantly focused on holistic scoring, yet rubric-based trait scoring provides richer diagnostic feedback for both students and instructors. While encoder-based models such as BERT have achieved competitive trait scoring performance, they are limited to score prediction and cannot generate the explanatory feedback that makes rubric-based assessment educationally useful. Large Language Models (LLMs), by contrast, naturally support both scoring and feedback generation within a single unified framework. However, most existing work on fine-tuned LLM-based rubric scoring relies on either large annotated corpora or synthetic data augmentation to achieve competitive performance, limiting practical adoption in settings where neither is readily available. This paper investigates whether a small, carefully balanced dataset of real essays can enable a fine-tuned LLM to achieve competitive rubric-based trait scoring. We fine-tune Qwen2.5-7B-Instruct using QLoRA on progressively augmented subsets of DREsSNew (371, 478, and 736 essays), targeting under-represented score bands at each stage, and propose a score-token weighted loss function that concentrates the training signal on score-bearing tokens. Data augmentation drives the majority of the performance improvement: expanding from 371 to 736 essays yields gains of +0.042 QWK on Content, +0.021 on Organization, +0.041 on Language, 0.024 on Total. Score-token weighting provides a further modest but consistent improvement, most notably on Content (+0.025). Our best configuration achieves per-trait QWK of 0.631 (Content), 0.612 (Organization), 0.625 (Language), and 0.669 (Total) on the DREsSNew test set using only 736 real essays - remaining competitive with models trained on datasets over 65 times larger while preserving the generative capability needed for downstream feedback generation.</description>
      <pubDate>Sat, 15 Aug 2026 00:00:00 GMT</pubDate>
      <guid isPermaLink="false">https://bura.brunel.ac.uk/handle/2438/34173</guid>
      <dc:date>2026-08-15T00:00:00Z</dc:date>
    </item>
    <item>
      <title>A transformer-based framework for device-free human activity recognition using Wi-Fi CSI</title>
      <link>https://bura.brunel.ac.uk/handle/2438/34169</link>
      <description>Title: A transformer-based framework for device-free human activity recognition using Wi-Fi CSI
Authors: Costa Nascimento dos Santos, Allan; Soares, Pamella; Galdino, Iandra; Ramos, Taiane C; Albuquerque, Célio; Guerra, Raphael; Fernandes, Natalia C; Muchaluat-Saade, Débora C; Ghinea, Gheorghita
Abstract: Human activity recognition (HAR) plays a pivotal role in ambient assisted living, particularly for monitoring the elderly and patients with chronic conditions. However, traditional approaches relying on wearable sensors or video cameras face significant challenges regarding user compliance and privacy intrusion. To mitigate these issues, this paper proposes a device-free sensing (DFS) ( https://github.com/mestrelan/MDA-CSI ) framework utilizing Wi-Fi channel state information (CSI), named named DFS-CSI. We introduce a robust Transformer-based architecture designed to capture long-range temporal dependencies in wireless signals. DFS-CSI was validated using a comprehensive dataset from 86 volunteers, ensuring high generalization capabilities across diverse human motion patterns.
Description: Data availability: &#xD;
The raw CSI data and the statistical social data from the dataset are available for research purposes and can be shared upon request.; Code availability: &#xD;
The Python code used for DFS-CSI is available at https://github.com/mestrelan/MDA-CSI .</description>
      <pubDate>Sat, 15 Aug 2026 00:00:00 GMT</pubDate>
      <guid isPermaLink="false">https://bura.brunel.ac.uk/handle/2438/34169</guid>
      <dc:date>2026-08-15T00:00:00Z</dc:date>
    </item>
  </channel>
</rss>

