Christos Bormpotsis

Dr. Christos Bormpotsis

orcid

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
15 September 2023 in Big Data and Cognitive Computing
Verified
Journal Article
Open access
PDF file
CBChristos Bormpotsis
Christos Bormpotsis
MSMohamed Sedky
Mohamed Sedky
APAsma Patel
Asma Patel
In the realm of foreign exchange (Forex) market predictions, Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) have been commonly employed. However, these models often exhibit instability due to vulnerability to data perturbations attributed to their monolithic architecture. Hence, this study proposes a novel neuroscience-informed modular network that harnesses closing prices and sentiments from Yahoo Finance and Twitter APIs. Compared to monolithic methods, the objective is to advance the effectiveness of predicting price fluctuations in Euro to British Pound Sterling (EUR/GBP). The proposed model offers a unique methodology based on a reinvigorated modular CNN, replacing pooling layers with orthogonal kernel initialisation RNNs coupled with Monte Carlo Dropout (MCoRNNMCD). It integrates two pivotal modules: a convolutional simple RNN and a convolutional Gated Recurrent Unit (GRU). These modules incorporate orthogonal kernel initialisation and Monte Carlo Dropout techniques to mitigate overfitting, assessing each module’s uncertainty. The synthesis of these parallel feature extraction modules culminates in a three-layer Artificial Neural Network (ANN) decision-making module. Established on objective metrics like the Mean Square Error (MSE), rigorous evaluation underscores the proposed MCoRNNMCD–ANN’s exceptional performance. MCoRNNMCD–ANN surpasses single CNNs, LSTMs, GRUs, and the state-of-the-art hybrid BiCuDNNLSTM, CLSTM, CNN–LSTM, and LSTM–GRU in predicting hourly EUR/GBP closing price fluctuations.

8 April 2025 in IEEE Transactions on Technology and Society
Verified
Journal Article
Closed access
CBChristos Bormpotsis
Christos Bormpotsis
MNMichael Nanos
Michael Nanos
APAsma Patel
Asma Patel
Financial decision-making, a cornerstone of individual prosperity and global economic stability, is hard to comprehend because it is a complex cognitive process concerned with emotional state and behavioural bias. This paper aims to decode the neural mechanisms behind financial behaviour to advance theoretical and empirical progress in neurofinance. Thus, to better capture financial behaviour, this article proposes an innovative framework that bridges neurofinance, neuroscience, and bio-inspired computational models, like the MCoRNNMCD-ANN. Key research areas include the role of neural processes driving decisions, the effect of cognitive preferences on judgment, and the potential of bio-inspired AI models to enhance understanding. The societal implications of this research seek to encourage equitable, stable and informed financial systems while addressing challenges at the intersection of neurofinance and neuroscience-informed AI.

18 August 2023 in Preprints.org
Verified
Preprint
Open access
PDF file
CBChristos Bormpotsis
Christos Bormpotsis
MSMohamed Sedky
Mohamed Sedky
APAsma Patel
Asma Patel
Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) have been utilised to forecast the foreign exchange market (Forex). However, such models usually exhibit unstable behaviour, as data perturbations often downgrade the functionality of the entire network due to their monolithic architecture. Hence, this study proposes a novel neuroscience-informed modular network applied to the closing prices and the sentiments retrieved from Yahoo Finance and Twitter APIs, aiming to anticipate better price fluctuations in Euro to British Pound Sterling (EUR/GBP) rather than monolithic methods. The proposed model is based on a new modular CNN, replacing pooling layers with orthogonal kernel initialisation RNNs coupled with Monte Carlo Dropout (MCD), namely MCoRNNMCD. It combines two modules: i) a convolutional simple RNN and ii) a convolutional Gated Recurrent Unit (GRU), where orthogonality and MCD are added to reduce the overfitting, assessing each module's uncertainty. These parallel feature extraction modules concatenate their outputs to a final three-layer Artificial Neural Network (ANN) decision-making module. A comprehensive comparison viewing objective evaluation metrics such as the Mean Square Error (MSE) proved that the proposed MCoRNNMCD-ANN outperformed single CNN, LSTM, GRU, and the state-of-the-art hybrid BiCuDNNLSTM, CLSTM, CNN-LSTM, and LSTM-GRU in forecasting hourly EUR/GBP closing price fluctuations.

23 August 2025 in Biomimetics
Verified
Journal Article
Open access
PDF file
CBChristos Bormpotsis
Christos Bormpotsis
MAMaria Anagnostouli
Maria Anagnostouli
MSMohamed Sedky
Mohamed Sedky
Eleni JelastopuluEleni Jelastopulu
Eleni Jelastopulu
APAsma Patel
Asma Patel
Anxiety and depression affect millions worldwide, yet stigma and long wait times often delay access to care. Mobile mental health apps can decrease these barriers by offering on-demand screening and support. Nevertheless, many machine and deep learning methods used in such tools perform poorly under severe class imbalance, yielding biased, poorly calibrated predictions. To address this challenge, this study proposes MCoG-LDPSNet, a brain-inspired model that combines dual, orthogonal encoding pathways with a novel Loss-Driven Parametric Swish (LDPS) activation. LDPS implements a neurobiologically motivated adaptive-gain mechanism via a learnable β parameter driven by calibration and confidence-aware loss signals that amplifies minority-class patterns while preserving overall reliability, enabling robust predictions under severe data imbalance. On a benchmark mental health corpus, MCoG-LDPSNet achieved AUROC = 0.9920 and G-mean = 0.9451, outperforming traditional baselines like GLMs, XGBoost, state-of-the-art deep models (CNN-BiLSTM-ATTN), and transformer-based approaches. After transfer learning to social media text, the MCoG-LDPSNet maintained a near-perfect AUROC of 0.9937. Integrated into the EmotiZen App with enhanced app features, MCoG-LDPSNet was associated with substantial symptom reductions (anxiety 28.2%; depression 42.1%). These findings indicate that MCoG-LDPSNet is an accurate, imbalance-aware solution suitable for scalable mobile screening of individuals for anxiety and depression.