Erçin Dinçer

Mr. Erçin Dinçer

orcid

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

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

Representative Publications

View All Publications
24 December 2025 in Symmetry
Verified
Journal Article
Open access
PDF file
Erçin DinçerErçin Dinçer
Erçin Dinçer
Zeynep Hilal KİLİMCİZeynep Hilal KİLİMCİ
Zeynep Hilal KİLİMCİ
This study introduces a unified and methodologically symmetric comparative framework for multivariate cryptocurrency forecasting, addressing long-standing inconsistencies in prior research where model families, feature sets, and preprocessing pipelines differ across studies. Under an identical and rigorously controlled experimental setup, we benchmark six deep learning architectures—LSTM, GPT-2, Informer, Autoformer, Temporal Fusion Transformer (TFT), and a Vanilla Transformer—together with four widely used econometric models (ARIMA, VAR, GARCH, and a Random Walk baseline). All models are evaluated using a shared multivariate feature space composed of more than forty technical indicators, identical normalization procedures, harmonized sliding-window formations, and aligned temporal splits across five high-liquidity assets (BTC, ETH, XRP, XLM, and SOL). The experimental results show that transformer-based architectures consistently outperform both the recurrent baseline and classical econometric models across all assets. This superiority arises from the ability of attention mechanisms to capture long-range temporal dependencies and adaptively weight informative time steps, whereas recurrent models suffer from vanishing-gradient limitations and restricted effective memory. The best-performing deep learning models achieve MAPE values of 0.0289 (BTC, GPT-2), 0.0198 (ETH, Autoformer), 0.0418 (XRP, Informer), 0.0469 (XLM, Informer), and 0.0578 (SOL, TFT), substantially improving upon the performance of both LSTM and all econometric baselines. These findings highlight the effectiveness of attention-based architectures in modeling volatility-driven nonlinear dynamics and establish a reproducible, symmetry-preserving benchmark for future research in deep-learning-based financial forecasting.

26 December 2025 in Preprints.org
Verified
Preprint
Open access
PDF file
Erçin DinçerErçin Dinçer
Erçin Dinçer
Zeynep Hilal KİLİMCİZeynep Hilal KİLİMCİ
Zeynep Hilal KİLİMCİ
Large Language Models (LLMs) have recently gained prominence for deployment on edge devices, owing to their potential to support privacy-preserving, low-latency, and offline inference. Nevertheless, their considerable computational and memory requirements present fundamental challenges in both real-time and offline scenarios. This systematic review synthesizes evidence from 49 studies, of which 40 were analyzed in depth, to investigate techniques, challenges, and applications of LLM deployment on edge devices. The studies were identified through a structured search and screening process, and data were extracted regarding model types, hardware platforms, optimization strategies, and performance outcomes. Findings indicate that hardware acceleration, model compression, and hybrid edge–cloud strategies can yield latency reductions of up to 972×, memory savings of up to 130×, and energy efficiency improvements exceeding 1600×, while largely preserving accuracy. Real-time deployments are predominantly applied in robotics, healthcare monitoring, and autonomous driving, whereas offline deployments are tailored to privacy-sensitive or batch-oriented contexts. The review also identifies persistent research gaps, including the absence of standardized benchmarks and the limited generalizability of results to real-world environments. It concludes by outlining future research directions, with particular emphasis on hardware–software co-design, federated learning, and secure task offloading.

27 November 2025 in Preprints.org
Verified
Preprint
Open access
PDF file
Erçin DinçerErçin Dinçer
Erçin Dinçer
Zeynep Hilal KİLİMCİZeynep Hilal KİLİMCİ
Zeynep Hilal KİLİMCİ
This study presents a comprehensive comparative analysis of recurrent and transformer-based deep learning architectures for multivariate cryptocurrency price forecasting. Five high-liquidity digital assets—Bitcoin (BTC), Ethereum (ETH), Ripple (XRP), Stellar (XLM), and Solana (SOL)—are modeled using an extensive technical indicator set generated through the pandas_ta library, incorporating trend-, momentum-, volatility-, and volume-based features. The experimental framework evaluates six architectures—LSTM, GPT-2, Informer, Autoformer, Temporal Fusion Transformer (TFT), and the Vanilla Transformer—under a unified preprocessing pipeline comprising data cleansing, missing-value imputation, normalization, and sliding-window sequence generation. Models are trained and tested using identical temporal partitions and optimization strategies to ensure methodological consistency. Forecasting performance is assessed through multiple error metrics, including MSE, MAE, RMSE, MAPE, and R2. The results indicate that transformer-based architectures generally outperform recurrent models in capturing long-range dependencies and complex feature interactions, particularly in multivariate settings with rich technical indicator inputs. Informer and Autoformer exhibit strong stability in longer horizons, whereas GPT-2 achieves competitive short-term accuracy despite its computational demands. Observed challenges include normalization inconsistencies, hyperparameter sensitivity, and significant training costs associated with large-scale transformer architectures. Overall, the findings highlight the potential of modern transformer-based approaches as robust and scalable alternatives for high-frequency cryptocurrency forecasting.

13 November 2026 in Biology
Verified
Journal Article
Open access
PDF file
AKAyhan Kucukmanisa
Ayhan Kucukmanisa
Zeynep Hilal KİLİMCİZeynep Hilal KİLİMCİ
Zeynep Hilal KİLİMCİ
RDRamazan Duvar
Ramazan Duvar
Erçin DinçerErçin Dinçer
Erçin Dinçer
High classification scores are meaningful only if the prediction target is physically consistent. We audited image-only, tabular-only and fusion experiments on 9000 ZooScan objects drawn from the 1,153,507-object PELGAS collection. The archived three-class labels were tertiles of the stored equivalent spherical diameter (ESD), not biomass classes. All 893,542 records exported with version v7.26 followed the documented ESD formula, whereas all 259,965 v7.24 records equalled twice the integer part of projected area divided by π. All 2017 v7.24 objects in the subset fell in the highest archived class. Recomputing ESD consistently from area and pixel calibration changed 3430 of the 9000 labels. The subset spans 13 cruises and 697 cruise–station groups and includes 2639 objects annotated as non-living. On the corrected target, tree-based controls using all descriptors reached a macro-F1 of about 0.998 under both object-stratified and station-grouped evaluation. In preliminary single-seed neural runs without area- or length-based descriptors, tabular networks reached a macro-F1 of 0.89, a Swin-T image model 0.946 and gated fusion 0.944; the image-based runs covered three folds and retained scale cues, so they do not isolate the contribution of each modality. Target provenance should be verified before multimodal performance rankings are used in ecological monitoring.

19 September 2026 in Mathematics
Verified
Journal Article
Open access
PDF file
Zeynep Hilal KİLİMCİZeynep Hilal KİLİMCİ
Zeynep Hilal KİLİMCİ
Erçin DinçerErçin Dinçer
Erçin Dinçer
Accurate cryptocurrency price forecasting represents an important yet challenging problem in financial time-series analysis due to the highly volatile, nonlinear, and noise-sensitive nature of digital asset markets. Although transformer-based architectures have recently demonstrated strong capabilities in temporal sequence modeling, their behavior under high-frequency cryptocurrency dynamics and the role of explainability-guided feature refinement remain insufficiently explored. To address this gap, this study presents a comprehensive transformer-based forecasting framework for hourly cryptocurrency price prediction and investigates the impact of explainability-guided feature optimization on forecasting performance, robustness, and interpretability. Five transformer architectures—Vanilla Transformer, Informer, Autoformer, Reformer, and Temporal Fusion Transformer (TFT)—are systematically evaluated across five major cryptocurrency assets: Bitcoin (BTC), Ethereum (ETH), Solana (SOL), Dogecoin (DOGE), and Ripple (XRP). The experimental framework employs Open, High, Low, Close, and Volume (OHLCV) data together with a broad set of engineered technical indicators and evaluates model performance using Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), Mean Squared Error (MSE), Mean Absolute Percentage Error (MAPE), and the coefficient of determination (R2). To improve interpretability and reduce feature redundancy, SHAP (SHapley Additive exPlanations) and LIME (Local Interpretable Model-agnostic Explanations) are integrated directly into the forecasting pipeline. Based on the resulting explanations, asset-specific feature subsets are constructed, and all models are subsequently retrained using the refined feature representations. The results show that explainability-guided feature refinement provides compact, model-aware, and interpretable feature subsets with competitive forecasting performance; however, its effect on prediction accuracy is dependent on the cryptocurrency asset, transformer architecture, retained feature subset size, and market conditions. Additional robustness, sensitivity, alternative feature-selection, and statistical significance analyses indicate that the SHAP–LIME Top-15 subset should be interpreted as a conservative dimensionality-reduction strategy rather than a universally optimal feature-selection rule. The findings further reveal that transformer architectures incorporating sparse attention, decomposition mechanisms, or gating structures generally provide stronger performance than the Vanilla Transformer under highly volatile hourly market conditions. Overall, the proposed framework demonstrates that combining transformer-based forecasting with explainability-guided feature refinement can support interpretable and parsimonious high-frequency financial time-series modeling, while highlighting the importance of evaluating robustness, feature-selection sensitivity, and statistical variability alongside average forecasting errors.