Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Preeti Patel has a total of 17 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Intimate partner violence and survivor-reported partner characteristics in Ghana, Kenya, and Tanzania

    by Preeti Patel, Sentirenla Longchar

    Intimate partner violence (IPV) remains a major global public health and humanrights challenge, with a disproportionate burden in sub-Saharan Africa. While women's experiences of IPV are well documented, fewer population-level studies examine survivor-reported partner characteristics and relational dynamics associated with perpetration. To estimate IPV prevalence in selected sub-Saharan African countries and examine demographic, socio-relational, attitudinal, and survivor-reported partnerstructural characteristics associated with IPV. We analysed data from the 2022-2023 Demographic and Health Surveys in Ghana, Kenya, and Tanzania (n = 59,458 ever-partnered women; 9593 reporting IPV). IPV prevalence was assessed in relation to demographic, socio-relational, and structural partnership characteristics. Random forest models were used togenerate descriptive rankings of the relative importance of selected survivor-reported partner structural characteristics across countries, with findings interpreted non-causally. IPV prevalence was highest in Kenya (21.28%; 95% CI: 20.71-21.87), followed by Ghana (14.26%; 95% CI: 13.56-15.02) and Tanzania (12.34%; 95% CI: 11.66-12.99). Partner economic status and occupation consistently ranked as important structural factors. Controlling behaviours were common among women experiencing IPV, with notable cross-country variation. Incorporating carefully designed partner-related indicators into population-based surveys can strengthen IPV surveillance, inform targeted prevention strategies, and support gender-transformative policy in sub-Saharan Africa.

  • Generative AI and organisational collective intelligence: A dependency-structured framework

    by Preeti Patel, Maitreyee Dey

    Generative AI (GenAI) is now part of everyday organisational knowledge work, affecting how information is gathered, interpreted, argued over, and stored. It should not be treated as a neutral aid: in many settings it can function as an active participant in collective intelligence, shaping what groups notice and how issues are framed and settled. This paper sets out a dependency-structured framework that connects information acquisition, sensemaking and framing, shared reasoning, coordinated action, and organisational memory. The framework is intended as a diagnostic and explanatory lens for organisational analysis, rather than a predictive or causal model. A key implication of the framework is that weaknesses in early stages can propagate upward and distort later decisions, even when outputs appear faster or more coherent. A hypothetical case of a mid-sized financial advisory firm illustrates how GenAI can strengthen performance while risking increased epistemic fragility when foundational processes are compressed or bypassed. The paper ends with diagnostic prompts and governance principles for information professionals, arguing that GenAI tends to amplify existing organisational tendencies rather than reliably augment intelligence.

  • Open Access

    SpectralNet-Enabled Root Cause Analysis of Frequency Anomalies in Solar Grids Using μPMU

    by Arnabi Modak, Maitreyee Dey, Preeti Patel, Soumya Prakash Rana

    The rapid integration of solar power into distribution grids has intensified challenges related to frequency instability caused by fluctuating renewable generation. These unexpected frequency variations are difficult to capture using traditional or supervised methods because they emerge from nonlinear, rapidly changing inverter grid interactions and often lack labelled examples. To address this, the present work introduces a unique, frequency-centric framework for unsupervised detection and root cause analysis of grid anomalies using high-resolution micro-Phasor Measurement Unit (μPMU) data. Unlike previous studies that focus primarily on voltage phasors or rely on predefined event labels, this work employs SpectralNet, a deep spectral clustering approach, integrated with autoencoder-based feature learning to model the nonlinear interactions between frequency, ROCOF, voltage, and current. These methods are particularly effective for unexpected frequency variations because they learn intrinsic, hidden structures directly from the data and can group abnormal frequency behavior without prior knowledge of event types. The proposed model autonomously identifies distinct root causes such as unbalanced loads, phase-specific faults, and phase imbalances behind hazardous frequency deviations. Experimental validation on a real solar-integrated distribution feeder in the UK demonstrates that the framework achieves superior cluster compactness and interpretability compared to traditional methods like K-Means, GMM, and Fuzzy C-Means. The findings highlight SpectralNet’s capability to uncover subtle, nonlinear patterns in μPMU data, offering an adaptive, data-driven tool for enhancing grid stability and situational awareness in renewable-rich power systems.

  • Open Access

    System Inertia Cost Forecasting Using Machine Learning: A Data-Driven Approach for Grid Energy Trading in Great Britain

    by Maitreyee Dey, Soumya Prakash Rana, Preeti Patel

    As modern power systems integrate more renewable and decentralised generation, maintaining grid stability has become increasingly challenging. This study proposes a data-driven machine learning framework for forecasting system inertia service costs—a key yet underexplored variable influencing energy trading and frequency stability in Great Britain. Using eight years (2017–2024) of National Energy System Operator (NESO) data, four models—Long Short-Term Memory (LSTM), Residual LSTM, eXtreme Gradient Boosting (XGBoost), and Light Gradient-Boosting Machine (LightGBM)—are comparatively analysed. LSTM-based models capture temporal dependencies, while ensemble methods effectively handle nonlinear feature relationships. Results demonstrate that LightGBM achieves the highest predictive accuracy, offering a robust method for inertia cost estimation and market intelligence. The framework contributes to strategic procurement planning and supports market design for a more resilient, cost-effective grid.

  • Open Access

    Transfer Learning-Based Distance-Adaptive Global Soft Biometrics Prediction in Surveillance

    by Sonjoy Ranjon Das, Henry Onilude, Bilal Hassan, Preeti Patel, Karim Ouazzane

    Soft biometric prediction—including age, gender, and ethnicity—is critical in surveillance applications, yet often suffers from performance degradation as the subject-to-camera distance increases. This study hypothesizes that embedding distance-awareness into the training process can mitigate such degradation and enhance model generalization across varying visual conditions. We propose a distance-adaptive, multi-task deep learning framework built upon EfficientNetB3, augmented with task-specific heads and trained progressively across four distance intervals (4 m to 10 m). A weighted composite loss function is employed to balance classification and regression objectives. The model is evaluated on a hybrid dataset combining the Front-View Gait (FVG) and MMV annotated pedestrian datasets, totaling over 19,000 samples. Experimental results demonstrate that the framework achieves up to 95% gender classification accuracy at 4 m and retains 85% accuracy at 10 m. Ethnicity prediction maintains an accuracy above 65%, while age estimation achieves a mean absolute error (MAE) ranging from 1.1 to 1.5 years. These findings validate the model’s robustness across distances and its superiority over conventional static learning approaches. Despite challenges such as computational overhead and annotation demands, the proposed approach offers a scalable and real-time-capable solution for distance-resilient biometric systems.


  • Open Access

    Enhancing Gender-Based Violence Research: Holistic Approaches to Data Collection and Analysis

    by Subeksha Shrestha, Preeti Patel, Sentirenla Longchar, Aiswarya Francis Xavier

    Gender-based violence (GBV) is a profound and pervasive societal issue, disproportionately affecting women across diverse settings, including homes, workplaces, and public spaces. Despite its prevalence, significant challenges impede research on GBV, particularly regarding data collection, analysis, and ethical handling. This study investigates the complexities inherent in GBV research, focusing on the obstacles posed by under-reporting, ethical considerations, data quality, and the need for cross-comparative standards. Using a combination of police records, web scraping, news reports, and survey data from USAID’s Demographic and Health Surveys (DHS), our study examines strategies to work with sensitive GBV datasets, while maintaining data integrity. Our study advocates for improved demographic surveying and data integration methodologies that can enhance data accuracy and comparability. The findings suggest that while technological advancements, particularly generative AI and machine learning approaches, offer promising avenues for automating survey processes, reducing costs, and enhancing data collection efficiency, they present the limitations of secondary datasets, a lack of data disaggregation, and discrepancies in data coding systems, which highlight the necessity of refining global data standards.


  • Open Access

    A data-centric approach to terminal unit’s fault categorization and optimal positioning in building HVAC systems using ensemble learning

    by Maitreyee Dey, Preeti Patel, Soumya Prakash Rana

    This paper focuses on the fault detection and diagnosis of terminal units (TUs) in a building located in London, utilizing real operational historical data to assess their performance and optimal placement across multiple floors. While precise locations of the TUs are unavailable, our method analyzes their operational behaviour for one month, applying popular machine learning models to detect and analyze faults effectively. By examining each TU individually and in the aggregate, we identify behavioural patterns that inform decisions regarding their positioning within the building. The dataset comprises over 2 million data points collected from 730 TUs, enabling a comprehensive analysis of their functionality and the impact of suboptimal thermostat placements. Our study employs three machine learning models-traditional multi-class Support Vector Machines and two ensemble methods: Random Forest (RF), and Adaptive Boosting (AdaBoost)-to classify TU behaviors into normal operation, heating faults, and cooling faults. Results indicate that RF outperforms the other models with an accuracy of 99.89%, while AdaBoost achieves an accuracy of 85% and SVM shows 47% accuracy. The findings underscore the potential of a data-driven approach to inform retrofitting decisions and enhance the reliability of HVAC systems. This research contributes valuable knowledge toward optimizing TU placement, ultimately leading to improved energy efficiency and indoor environmental quality.