arXiv subjects

Dimitris Askounis

Publications and source records attributed to Dimitris Askounis.

28 records · Page 2Linked to original sources

arXiv2023

Targeted demand response for flexible energy communities using clustering techniques

The present study proposes clustering techniques for designing demand response (DR) programs for commercial and residential prosumers. The goal is to alter the consumption behavior of the prosumers within a distributed energy community in Italy. This aggregation aims to: a) minimize the reverse power flow at the primary substation, occuring when generation from solar panels in the local grid exceeds consumption, and b) shift the system wide peak demand, that typically occurs during late afternoon. Regarding the clustering stage, we consider daily prosumer load profiles and divide them across the extracted clusters. Three popular machine learning algorithms are employed, namely k-means, k-medoids and agglomerative clustering. We evaluate the methods using multiple metrics including a novel metric proposed within this study, namely peak performance score (PPS). The k-means algorithm with dynamic time warping distance considering 14 clusters exhibits the highest performance with a PPS of 0.689. Subsequently, we analyze each extracted cluster with respect to load shape, entropy, and load types. These characteristics are used to distinguish the clusters that have the potential to serve the optimization objectives by matching them to proper DR schemes including time of use, critical peak pricing, and real-time pricing. Our results confirm the effectiveness of the proposed clustering algorithm in generating meaningful flexibility clusters, while the derived DR pricing policy encourages consumption during off-peak hours. The developed methodology is robust to the low availability and quality of training datasets and can be used by aggregator companies for segmenting energy communities and developing personalized DR policies.

cs.LG↗

arXiv2023

A comparative assessment of deep learning models for day-ahead load forecasting: Investigating key accuracy drivers

Short-term load forecasting (STLF) is vital for the effective and economic operation of power grids and energy markets. However, the non-linearity and non-stationarity of electricity demand as well as its dependency on various external factors renders STLF a challenging task. To that end, several deep learning models have been proposed in the literature for STLF, reporting promising results. In order to evaluate the accuracy of said models in day-ahead forecasting settings, in this paper we focus on the national net aggregated STLF of Portugal and conduct a comparative study considering a set of indicative, well-established deep autoregressive models, namely multi-layer perceptrons (MLP), long short-term memory networks (LSTM), neural basis expansion coefficient analysis (N-BEATS), temporal convolutional networks (TCN), and temporal fusion transformers (TFT). Moreover, we identify factors that significantly affect the demand and investigate their impact on the accuracy of each model. Our results suggest that N-BEATS consistently outperforms the rest of the examined models. MLP follows, providing further evidence towards the use of feed-forward networks over relatively more sophisticated architectures. Finally, certain calendar and weather features like the hour of the day and the temperature are identified as key accuracy drivers, providing insights regarding the forecasting approach that should be used per case.

cs.LG↗

arXiv2023

Context-aware attention layers coupled with optimal transport domain adaptation and multimodal fusion methods for recognizing dementia from spontaneous speech

Alzheimer's disease (AD) constitutes a complex neurocognitive disease and is the main cause of dementia. Although many studies have been proposed targeting at diagnosing dementia through spontaneous speech, there are still limitations. Existing state-of-the-art approaches, which propose multimodal methods, train separately language and acoustic models, employ majority-vote approaches, and concatenate the representations of the different modalities either at the input level, i.e., early fusion, or during training. Also, some of them employ self-attention layers, which calculate the dependencies between representations without considering the contextual information. In addition, no prior work has taken into consideration the model calibration. To address these limitations, we propose some new methods for detecting AD patients, which capture the intra- and cross-modal interactions. First, we convert the audio files into log-Mel spectrograms, their delta, and delta-delta and create in this way an image per audio file consisting of three channels. Next, we pass each transcript and image through BERT and DeiT models respectively. After that, context-based self-attention layers, self-attention layers with a gate model, and optimal transport domain adaptation methods are employed for capturing the intra- and inter-modal interactions. Finally, we exploit two methods for fusing the self and cross-attention features. For taking into account the model calibration, we apply label smoothing. We use both performance and calibration metrics. Experiments conducted on the ADReSS and ADReSSo Challenge datasets indicate the efficacy of our introduced approaches over existing research initiatives with our best performing model reaching Accuracy and F1-score up to 91.25% and 91.06% respectively.

cs.CL↗

arXiv2023

Calibration of Transformer-based Models for Identifying Stress and Depression in Social Media

In today's fast-paced world, the rates of stress and depression present a surge. Social media provide assistance for the early detection of mental health conditions. Existing methods mainly introduce feature extraction approaches and train shallow machine learning classifiers. Other researches use deep neural networks or transformers. Despite the fact that transformer-based models achieve noticeable improvements, they cannot often capture rich factual knowledge. Although there have been proposed a number of studies aiming to enhance the pretrained transformer-based models with extra information or additional modalities, no prior work has exploited these modifications for detecting stress and depression through social media. In addition, although the reliability of a machine learning model's confidence in its predictions is critical for high-risk applications, there is no prior work taken into consideration the model calibration. To resolve the above issues, we present the first study in the task of depression and stress detection in social media, which injects extra linguistic information in transformer-based models, namely BERT and MentalBERT. Specifically, the proposed approach employs a Multimodal Adaptation Gate for creating the combined embeddings, which are given as input to a BERT (or MentalBERT) model. For taking into account the model calibration, we apply label smoothing. We test our proposed approaches in three publicly available datasets and demonstrate that the integration of linguistic features into transformer-based models presents a surge in the performance. Also, the usage of label smoothing contributes to both the improvement of the model's performance and the calibration of the model. We finally perform a linguistic analysis of the posts and show differences in language between stressful and non-stressful texts, as well as depressive and non-depressive posts.

cs.CL↗

arXiv2023

Neural Architecture Search with Multimodal Fusion Methods for Diagnosing Dementia

Alzheimer's dementia (AD) affects memory, thinking, and language, deteriorating person's life. An early diagnosis is very important as it enables the person to receive medical help and ensure quality of life. Therefore, leveraging spontaneous speech in conjunction with machine learning methods for recognizing AD patients has emerged into a hot topic. Most of the previous works employ Convolutional Neural Networks (CNNs), to process the input signal. However, finding a CNN architecture is a time-consuming process and requires domain expertise. Moreover, the researchers introduce early and late fusion approaches for fusing different modalities or concatenate the representations of the different modalities during training, thus the inter-modal interactions are not captured. To tackle these limitations, first we exploit a Neural Architecture Search (NAS) method to automatically find a high performing CNN architecture. Next, we exploit several fusion methods, including Multimodal Factorized Bilinear Pooling and Tucker Decomposition, to combine both speech and text modalities. To the best of our knowledge, there is no prior work exploiting a NAS approach and these fusion methods in the task of dementia detection from spontaneous speech. We perform extensive experiments on the ADReSS Challenge dataset and show the effectiveness of our approach over state-of-the-art methods.

cs.LG↗

arXiv2023

Detecting Dementia from Speech and Transcripts using Transformers

Alzheimer's disease (AD) constitutes a neurodegenerative disease with serious consequences to peoples' everyday lives, if it is not diagnosed early since there is no available cure. Alzheimer's is the most common cause of dementia, which constitutes a general term for loss of memory. Due to the fact that dementia affects speech, existing research initiatives focus on detecting dementia from spontaneous speech. However, little work has been done regarding the conversion of speech data to Log-Mel spectrograms and Mel-frequency cepstral coefficients (MFCCs) and the usage of pretrained models. Concurrently, little work has been done in terms of both the usage of transformer networks and the way the two modalities, i.e., speech and transcripts, are combined in a single neural network. To address these limitations, first we represent speech signal as an image and employ several pretrained models, with Vision Transformer (ViT) achieving the highest evaluation results. Secondly, we propose multimodal models. More specifically, our introduced models include Gated Multimodal Unit in order to control the influence of each modality towards the final classification and crossmodal attention so as to capture in an effective way the relationships between the two modalities. Extensive experiments conducted on the ADReSS Challenge dataset demonstrate the effectiveness of the proposed models and their superiority over state-of-the-art approaches.

cs.CL↗

arXiv2022

A Multimodal Approach for Dementia Detection from Spontaneous Speech with Tensor Fusion Layer

Alzheimer's disease (AD) is a progressive neurological disorder, meaning that the symptoms develop gradually throughout the years. It is also the main cause of dementia, which affects memory, thinking skills, and mental abilities. Nowadays, researchers have moved their interest towards AD detection from spontaneous speech, since it constitutes a time-effective procedure. However, existing state-of-the-art works proposing multimodal approaches do not take into consideration the inter- and intra-modal interactions and propose early and late fusion approaches. To tackle these limitations, we propose deep neural networks, which can be trained in an end-to-end trainable way and capture the inter- and intra-modal interactions. Firstly, each audio file is converted to an image consisting of three channels, i.e., log-Mel spectrogram, delta, and delta-delta. Next, each transcript is passed through a BERT model followed by a gated self-attention layer. Similarly, each image is passed through a Swin Transformer followed by an independent gated self-attention layer. Acoustic features are extracted also from each audio file. Finally, the representation vectors from the different modalities are fed to a tensor fusion layer for capturing the inter-modal interactions. Extensive experiments conducted on the ADReSS Challenge dataset indicate that our introduced approaches obtain valuable advantages over existing research initiatives reaching Accuracy and F1-score up to 86.25% and 85.48% respectively.

cs.CL↗

arXiv2022

Comparison of Missing Data Imputation Methods using the Framingham Heart study dataset

Cardiovascular disease (CVD) is a class of diseases that involve the heart or blood vessels and according to World Health Organization is the leading cause of death worldwide. EHR data regarding this case, as well as medical cases in general, contain missing values very frequently. The percentage of missingness may vary and is linked with instrument errors, manual data entry procedures, etc. Even though the missing rate is usually significant, in many cases the missing value imputation part is handled poorly either with case-deletion or with simple statistical approaches such as mode and median imputation. These methods are known to introduce significant bias, since they do not account for the relationships between the dataset's variables. Within the medical framework, many datasets consist of lab tests or patient medical tests, where these relationships are present and strong. To address these limitations, in this paper we test and modify state-of-the-art missing value imputation methods based on Generative Adversarial Networks (GANs) and Autoencoders. The evaluation is accomplished for both the tasks of data imputation and post-imputation prediction. Regarding the imputation task, we achieve improvements of 0.20, 7.00% in normalised Root Mean Squared Error (RMSE) and Area Under the Receiver Operating Characteristic Curve (AUROC) respectively. In terms of the post-imputation prediction task, our models outperform the standard approaches by 2.50% in F1-score.

cs.LG↗

arXiv2022

Explainable Identification of Dementia from Transcripts using Transformer Networks

Alzheimer's disease (AD) is the main cause of dementia which is accompanied by loss of memory and may lead to severe consequences in peoples' everyday life if not diagnosed on time. Very few works have exploited transformer-based networks and despite the high accuracy achieved, little work has been done in terms of model interpretability. In addition, although Mini-Mental State Exam (MMSE) scores are inextricably linked with the identification of dementia, research works face the task of dementia identification and the task of the prediction of MMSE scores as two separate tasks. In order to address these limitations, we employ several transformer-based models, with BERT achieving the highest accuracy accounting for 87.50%. Concurrently, we propose an interpretable method to detect AD patients based on siamese networks reaching accuracy up to 83.75%. Next, we introduce two multi-task learning models, where the main task refers to the identification of dementia (binary classification), while the auxiliary one corresponds to the identification of the severity of dementia (multiclass classification). Our model obtains accuracy equal to 86.25% on the detection of AD patients in the multi-task learning setting. Finally, we present some new methods to identify the linguistic patterns used by AD patients and non-AD ones, including text statistics, vocabulary uniqueness, word usage, correlations via a detailed linguistic analysis, and explainability techniques (LIME). Findings indicate significant differences in language between AD and non-AD patients.

cs.CL↗

arXiv2021

Designing a Cyber-security Culture Assessment Survey Targeting Critical Infrastructures During Covid-19 Crisis

The paper at hand presents the design of a survey aiming at the cyber-security culture assessment of critical infrastructures during the COVID-19 crisis, when living reality was heavily disturbed and working conditions fundamentally affected. The survey is rooted in a security culture framework layered into two levels, organizational and individual, further analyzed into 10 different security dimensions consisted of 52 domains. An in-depth questionnaire building analysis is presented focusing on the aims, goals, and expected results. It concludes with the survey implementation approach while underlining the framework's first application and its revealing insights during a global crisis.

cs.CR↗