Adaptation Method Research Articles

Stylized Face Sketch Extraction via Generative Prior with Limited Data

AbstractFacial sketches are both a concise way of showing the identity of a person and a means to express artistic intention. While a few techniques have recently emerged that allow sketches to be extracted in different styles, they typically rely on a large amount of data that is difficult to obtain. Here, we propose StyleSketch, a method for extracting high‐resolution stylized sketches from a face image. Using the rich semantics of the deep features from a pretrained StyleGAN, we are able to train a sketch generator with 16 pairs of face and the corresponding sketch images. The sketch generator utilizes part‐based losses with two‐stage learning for fast convergence during training for high‐quality sketch extraction. Through a set of comparisons, we show that StyleSketch outperforms existing state‐of‐the‐art sketch extraction methods and few‐shot image adaptation methods for the task of extracting high‐resolution abstract face sketches. We further demonstrate the versatility of StyleSketch by extending its use to other domains and explore the possibility of semantic editing. The project page can be found in https://kwanyun.github.io/stylesketch_project.

Feature disentanglement based domain adaptation network for cross-scene coastal wetland hyperspectral image classification

At present, domain adaptation (DA) methods have made noteworthy advancements in cross-scene hyperspectral image (HSI) classification. Their success largely hinges on the alignment of distributions between source and target domains, which is a critical step in extracting domain-invariant features. However, this intense focus on domain-invariant feature extraction frequently leads to the neglect of class-discriminative features, limiting their utility in cross-scene coastal wetland classification, where the nuanced identification of different classes is crucial. In this paper, a feature disentanglement based domain adaptation network (FDDAN) is proposed to disentangle and exclude domain-specific features and class-invariant features, thereby obtaining class-specific domain-invariant features for classification tasks. Specifically, a transformer and convolution fusion-based feature extraction network is designed to capture global–local mixed features. To align domain distributions and learn shared features, two corresponding disentanglers separate domain-invariant features and domain-specific features from mixed features. Furthermore, to allow domain-invariant features containing purer category discriminative information, class-invariant features are also segregated. In addition, an adversarial learning strategy between three features is utilized to simultaneously enhance the transferability and discriminability of domain-invariant features. The effectiveness of FDDAN is demonstrated by the experimental results obtained from three cross-scene unmanned aerial vehicle datasets collected in coastal wetlands and a publicly available dataset. We will make our datasets and source code publicly available upon acceptance of the paper.

THE DECLINE OF INFORMATION RESISTANCE AS A PREREQUISITE FOR THE PRACTICE OF POLITICAL MANIPULATION

In this publication, based on the behavioral approach, conceptualization and discursive method, a study is carried out of the possibilities and characteristics of the influence of an objectively developing information array, distributed in accordance with modern communication trends, on the cognitive and emotional processes of the mentality of consumers of an information product in general and on the function of information resistance in particular. As a hypothesis, a statement is formulated according to which, in conditions of objective growth of generated information and the introduction of information (digital, network) addictions into the psyche, in combination with the inertia of perception inherent in the human brain, information resistance decreases, which increases the likelihood of the successful use of manipulative practices as a variation of suggestive communication. A definition of information resistance is given as resistance to the effects of unverified and/or unreliable information. The determinants of the vector of falling information resistance are revealed and studied, such as inertia of perception, pandemic dispersion of digital dependence, information overload. The validity of the correlation between the rate of decrease in information resistance and the effective dispersion of manipulations is substantiated. The leading varieties of modern manipulative practices, which are generated, among other things, by a decrease in information resistance, are characterized - the spread of deepfakes, the synthesis of cybersimulacra (including astroturfing and sockpuppeting), the imposition of behavioral patterns of reference groups, the cultivation and popularization of ersatz culture. Based on the results of the study, a position is formulated according to which a decrease in the level of information resistance, associated with a decline in the intellectual functions of human consciousness, activates primitive archaic forms of consciousness. It is noted that due to the reduction of cognitive methods of adaptation to the surrounding world to the simplest irrational forms, space opens up for the implementation of power relations through emotionally charged manipulations.

CONCEPTUAL PROBLEMS WITH DISEMBODIED COGNITION IN LEARNING ENVIRONMENT(S) AND THE ALTERNATIVE OF EMBODIED CREATIVITY

The article is motivated by today’s practical realities and theoretical transformations that have affected education on a major scale. The shift from in-person to online classrooms at the beginning of the pandemic brought forward a series of issues related to embodied conditioning for creative strategies in the learning process. After presenting the shortcomings of the disembodied approach to education, we emphasize the role of the embodied (somatic, motoric, affective) aspects of education and discuss the embodied skills of creativity in a variety of learning environments. While so called embodied creativity became a fast-developing field due to mostly quantitative experiments in teaching-learning environments, it still lacks some conceptual clarification, especially in relation to its genesis in the paradigm of embodied cognition. Hence the main goal of this conceptual article is to extend, through the method of theory adaptation, the existing body of research on embodied cognition in academic environments to show how the embodied teaching and learning paradigm presents embodied creativity methods as an alternative to the disembodied approach to education and how technological environments provide an opportunity for such purposes.

A Deterministic Calibration Method for the Thermodynamic Model of Gas Turbines

Performance adaptation is an effective way to improve the accuracy of gas turbine performance models. Although current performance adaptation methods, such as those using genetic algorithms or evolutionary computation to modify component characteristic maps, are useful for finding good solutions, they are essentially searching methods and suffer from long computation time. This paper presents a novel approach that can achieve good performance adaptation with low time complexity and without using any searching method. In this method, the actual component performance parameters are first estimated using engine measurements at different operating conditions. For each operating condition, some scaling factors are introduced and calculated to indicate the difference between the actual and predicted component performance parameters. Afterward, an interpolating algorithm is adopted to synthesize the scaling factors for modifying all major component maps. The adapted component maps are then able to make the engine model match all the gas path measurements and achieve the required accuracy of the engine performance model. The proposed approach has been tested with a model high-bypass turbofan engine using simulated data. The results show that the proposed performance adaptation approach can effectively improve the model’s accuracy. Specifically, the prediction errors can be reduced from about 9% to about 0.6%. In addition, this approach has much less computational complexity compared to other optimization-based counterparts.

Advanced Sparse Filtering-Based Domain Adaptation for Fault Diagnosis in Variable Working Conditions

Traditional domain adaptation (DA) methods often encounter challenges with cross-domain feature extraction and the precise assessment of domain differences. To overcome these limitations, we introduce the Enhanced Sparse Filtering-Based Domain Adaptation (ESFBDA) method. This method distinguishes itself by enhancing sparse filtering (SF) with the integration of row-column normalization and a cosine penalty, specifically designed to minimize feature loss—a critical issue in existing DA techniques. Additionally, we employ Bootstrap resampling to refine domain distribution alignment, a novel step that boosts feature similarity and effectiveness in DA. This integrated approach ensures more accurate feature extraction, which is crucial for the classifier's fault detection capability. In our study, through two distinct experiments on Electro-Hydrostatic Actuator (EHA) internal leakage and bearing fault diagnosis, the ESFBDA method demonstrated remarkable accuracy, significantly surpassing traditional approaches and showcasing its robust applicability across varied diagnostic scenarios.

Constrained multitasking optimization via co-evolution and domain adaptation

Constrained multitasking optimization (CMTO) obtains increasing attention recently. The goal of CMTO is to handle multiple constrained tasks simultaneously. There are two limitations of existing studies on CMTO: (i) existing knowledge transfer techniques for CMTO may not work due to non-intersecting feasible domains and the unconsidered relationship between constraints and objectives; and (ii) knowledge diversity is lacking in existing CMTO algorithms because the representation of knowledge is biased to feasible solutions. To address these limitations, this paper proposes a co-evolution and domain adaptation (CEDA) method for CMTO. First, a new constraint relaxation-based domain adaptation technique for knowledge transfer is devised. Domain adaptation can effectively address the limitations imposed by non-intersecting feasible domains. In addition, as the population evolves, the knowledge representation is biased to different kinds of solutions. Second, a co-evolutionary strategy is proposed to improve the knowledge diversity. The two proposed techniques are with generality and can be readily integrated into different multitasking frameworks. In this paper, the CEDA method is combined with two popular multitasking (i.e., multifactorial-based and multi-population-based) frameworks. The constructed CEDA-based algorithms are compared with fifteen state-of-the-art algorithms on a CMTO benchmark suite and a real-world application. Experimental results demonstrate the superiority of the proposed CEDA method.

Physical Activity as a Means of Adaptation of People with Impaired Health

The article examines the benefits of physical activity for people with impaired health using the example of cancer patients. The features of the use of swimming and running for the rehabilitation of patients are considered. The purpose of the article is to highlight the advantages and disadvantages of the considered types of physical activity for people with impaired health. The swimming study was conducted on an experimental group formed on the basis of the NGO “Together against Cancer” (Krasnoyarsk). The author has developed training conditions, created methods for dynamic adaptation of physical activity in the form of exercises, training distances, theoretical approaches have been confirmed experimentally. The analysis of the test results was carried out and the structuring of the collected material was performed. To conduct research on the impact of running on rehabilitation processes, a theoretical analysis of scientific, psychological, pedagogical, sociological and normative literature was performed and specific examples of participants in competitions were considered. Running does not require the study of special techniques, but an initial level of physical fitness is required to enter the training process, which is excluded if the adaptive swimming technique developed by the author is used. When using it, the set of muscles involved in the training process expands, which has a greater restorative effect on the entire body. Swimming is therefore more suitable for people with poor health and is often the only acceptable form of physical activity.

Reinforced fuzzy domain adaptation: Revolutionizing data-unaccessible rotating machinery fault diagnosis across multiple domains

Existing domain adaptation methods typically require access to raw data from both the source and target domains, increasing the risk of data leakage. In light of this, a fuzzy domain adaptation-based rotation machinery fault diagnosis method that does not require access to data is proposed in this paper. Specifically, during the source domain training phase, all source models are simultaneously trained by contributing model parameters. The developed joint training strategy significantly enhances the multi-domain performance of each private source model. Simultaneously, a model incorporating a fusion fuzzy strategy is constructed to acquire fuzzy rules for the source domain samples and anchor points corresponding to each fault mode. During the target domain adaptation phase, firstly, under the setting of frozen fuzzy rules, the source feature extractor is retrained using target data. Subsequently, an anchor point alignment strategy is introduced to transfer diagnostic knowledge between domains. Finally, a filtering strategy is devised to obtain high-confidence pseudo-labels for the target domain, coupled with a self-supervised learning strategy for the final optimization of the target model. Experimental results demonstrate significant achievements of the proposed method across twelve multi-source cross-domain fault diagnosis cases, with an average diagnostic accuracy of 98.88%. Encouragingly, the proposed method maintains outstanding performance even without accessing source data, outperforming several other methods.

Reject inference in credit scoring based on cost-sensitive learning and joint distribution adaptation method

As traditional credit evaluation methods generally only use accepted sample modeling, the rejected data is omitted, which means the model's prediction of new customers is biased. However, reject inference can be used to solve this credit evaluation sample selection bias. This paper proposes a new reject inference method based on joint distribution adaptation (JDA) and cost-sensitive semi-supervised support vector machines (CS4VM). First, this method uses both accepted (labeled) samples and rejected (unlabeled) samples modeling, which overcomes the deviations in traditional credit evaluation methods. Second, as the accepted sample and the rejected sample distributions are different, this method reduces the distribution differences between the accepted and rejected sample sets, which ensures that the sample data conforms to the basic assumptions in the semi-supervised model, and improves the performance of the classification model. Third, this method reduces the overall cost in the actual credit business by considering both the traditional misclassification costs when mining the default samples and the different decision weights for the accepted and rejected samples. Finally, an empirical study verifies the excellent predictive performance of the proposed method and effectively reduces the total credit costs.

Probabilistic assessment of scalar transport under hydrodynamically unstable flows in heterogeneous porous media

Quantitative predictions of scalar transport in natural porous media is a nontrivial task given the presence of multi-scale spatial heterogeneity in the permeability field. Due to data scarcity, the structural map of the permeability field is subject to uncertainty and therefore, model predictions are uncertain. For such reasons, probabilistic models of flow and transport in natural porous media are required in risk assessment and to provide reliable decision making under uncertainty. Further complexities arise when the viscosity of the injected solute differs from that of the ambient fluid. Under the presence of viscosity contrast, hydrodynamic instabilities give rise to viscous fingering, which induces additional disorder in both velocity and solute concentration fields. This work examines the combined role of viscous fingering and permeability heterogeneity in the probabilistic description of transport predictions. In particular, we focus on metrics that are important for risk analysis, such as the solute plume’s early arrival times and the maximum concentration observed at a given location. We propose to use the Projection Pursuit Adaptation (PPA) method in the Polynomial Chaos Expansion (PCE) framework to quantify uncertainty in transport model predictions. The PPA method is a data-driven approach that optimally represents a given quantity of interest in a low-dimensional manifold. Unlike other dimension reduction techniques in uncertainty quantification, the PPA method utilizes non-linear information of the quantity of interest to identify the low-dimensional manifold, thereby increasing the likelihood of finding a more accurate lower-dimensional space. Moreover, the PPA model converges to the physical solution in a mean squared sense with respect to the polynomial order, enabling the construction of a converged model even with limited available data. Then, the PPA results are compared to Monte Carlo simulations using the same amount of data. This comparison illustrates that while Monte Carlo simulations are able to capture low-order statistics, they struggle to represent more detailed probability density functions. Our results show how the combined effect of permeability heterogeneity and viscosity contrast can enhance the mobility of the solute plume.

Diatom-Inspired Structural Adaptation According to Mode Shapes: A Study on 3D Structures and Software Tools.

Diatoms captivate both biologists and engineers with their remarkable mechanical properties and lightweight design principles inherent in their shells. Recent studies have indicated that diatom frustules possess optimized shapes that align with vibrational modes, suggesting an inherent adaptation to vibratory loads. The mode shape adaptation method is known to significantly alter eigenfrequencies of 1D and 2D structures to prevent undesired vibration amplitudes. Leveraging this insight, the diatom-inspired approach to deform structures according to mode shapes was extended to different complex 3D structures, demonstrating a significant enhancement in eigenfrequencies with distinct mode shapes. Through extensive parameter studies, frequency increases exceeding 200% were obtained, showcasing the method's effectiveness. In the second study part, the studied method was integrated into a user-friendly, low-code software facilitating swift and automated structural adjustments for eigenfrequency optimization. The created software tools, encompassing various components, were successfully tested on the example structures demonstrating the versatility and practicality of implementing biomimetic strategies in engineering designs. Thus, the present investigation does not only highlight the noteworthiness of the structural adaptation method inspired by diatoms in maximizing eigenfrequencies, but also originate software tools permitting different users to easily apply the method to distinct structures that have to be optimized, e.g., lightweight structures in the mobility or aerospace industry that are susceptible toward vibrations.

Single imbalanced domain generalization network for intelligent fault diagnosis of compressors in HVAC systems under unseen working conditions

Effective fault diagnosis of compressors in heating, ventilation, and air conditioning (HVAC) systems is critical to ensure service reliability and boost energy efficiency. HVAC compressors are distributed in different areas and work under heterogeneous conditions, which poses emerging challenges to their data-driven modeling. Most existing methods assume multiple category-balanced source domains for model training. Although domain adaptation and generalization methods have emerged to address the data distribution discrepancies in cross-domain fault diagnosis, limited source domains and imbalanced fault categories across domains still constrain the real-world applicability of data-driven models in HVAC compressor fault diagnosis under unseen working conditions. Therefore, this paper studies a significant fault diagnosis problem named single imbalanced domain generalization (SIDG) and proposes a corresponding network (SIDGNet) for intelligent HVAC compressor fault diagnosis. Specifically, a rare fault diagnosis module combined with focal loss is introduced to tackle the class imbalance problem. To achieve better diagnostic boundaries and resist unknown data distribution discrepancies, joint supervised contrastive learning and adversarial learning with specialized data augmentation are introduced as auxiliary modules to improve the robustness and generalizability of SIDGNet. An improved uncertainty-based dynamic weighting mechanism is developed to intelligently balance the weights of module-specific losses during training, which ensures an efficient and stable optimization process. Extensive SIDG fault diagnosis experiments conducted on HVAC compressors demonstrate the superiority of SIDGNet over existing models in SIDG scenarios.

Information filtering and interpolating for semi-supervised graph domain adaptation

Graph domain adaptation, which falls under the umbrella of graph transfer learning, involves transferring knowledge from a labeled source graph to improve prediction accuracy on an unlabeled target graph, where both graphs have identical label spaces but exhibit distribution discrepancies due to temporal data shifts or distinct data collection methods. This adaptation is complicated by the challenges of graph-specific domain discrepancies and cross-graph label scarcity. This paper proposes a semi-supervised Graph domain adaptation method via Information Filtering and Interpolating (GIFI). Specifically, GIFI utilizes a parameterized graph reduction module and a variational information bottleneck to adequately filter out irrelevant information from the source and target graphs to eliminate distribution discrepancy. GIFI also introduces an interpolation-enhanced pseudo-labeling strategy for cross-graph semi-supervised learning, which can mitigate model over-fitting on domain-specific features and limited labeled nodes, thus improving the model’s adaptation and discriminative capability. Experimental results on various graph domain adaptation benchmarks demonstrate GIFI’s superior performance over state-of-the-art methods. Our code is available at https://github.com/joe817/GIFI.

Bridging the Divide: Indigenous Language Integration in Open Educational Resource Accessibility

The use of information resources is dependent on their accessibility by users. The accessibility of information has attracted the attention of researchers from different parts of the world. However, Open Educational Resources (OERs) accessibility issues have not been adequately addressed in South Africa. This article explores the use of indigenous languages in the development and adaptation of OERs for accessibility at the University of South Africa. A total of 108 OERs were available for selection on UnisaIR. Ten of the 108 OERs were purposefully selected. Exploring and analysing OER development at Unisa took a qualitative and exploratory approach. This article presents the findings of a thematic analysis that was used to identify dominant themes in OER. The findings highlight two main concerns that underpin OER operations at Unisa: 1) translation as a method of OER creation and/or adaptation; and 2) deficiency of OER creation in indigenous languages. Though using indigenous languages in the construction of OERs can help to address the accessibility gap in the supply of teaching and learning resources, this aspect of accessibility as a discourse in the OER space has yet to be explored extensively. It is a matter of concern that many OERs at Unisa are written only in English. To address the gap in accessibility of OERs in the South African context, the following three language-related factors should support the creation of OERs: development in local languages, linguistic pluralism, and translation into indigenous languages.

Can surgical computer vision benefit from large-scale visual foundation models?

We investigate whether foundation models pretrained on diverse visual data could be beneficial to surgical computer vision. We use instrument and uterus segmentation in mini-invasive procedures as benchmarks. We propose multiple supervised, unsupervised and few-shot supervised adaptations of foundation models, including two novel adaptation methods. We use DINOv1, DINOv2, DINOv2 with registers, and SAM backbones, with the ART-Net surgical instrument and the SurgAI3.8K uterus segmentation datasets. We investigate five approaches: DINO unsupervised, few-shot learning with a linear decoder, supervised learning with the proposed DINO-UNet adaptation, DPT with DINO encoder, and unsupervised learning with the proposed SAM adaptation. We evaluate 17 models for instrument segmentation and 7 models for uterus segmentation and compare to existing ad hoc models for the tasks at hand. We show that the linear decoder can be learned with few shots. The unsupervised and linear decoder methods obtain slightly subpar results but could be considered useful in data scarcity settings. The unsupervised SAM model produces finer edges but has inconsistent outputs. However, DPT and DINO-UNet obtain strikingly good results, defining a new state of the art by outperforming the previous-best by 5.6 and 4.1 pp for instrument and 4.4 and 1.5 pp for uterus segmentation. Both methods obtain semantic and spatial precision, accurately segmenting intricate details. Our results show the huge potential of using DINO and SAM for surgical computer vision, indicating a promising role for visual foundation models in medical image analysis, particularly in scenarios with limited or complex data.

A Multi-Modal Egocentric Activity Recognition Approach towards Video Domain Generalization.

Egocentric activity recognition is a prominent computer vision task that is based on the use of wearable cameras. Since egocentric videos are captured through the perspective of the person wearing the camera, her/his body motions severely complicate the video content, imposing several challenges. In this work we propose a novel approach for domain-generalized egocentric human activity recognition. Typical approaches use a large amount of training data, aiming to cover all possible variants of each action. Moreover, several recent approaches have attempted to handle discrepancies between domains with a variety of costly and mostly unsupervised domain adaptation methods. In our approach we show that through simple manipulation of available source domain data and with minor involvement from the target domain, we are able to produce robust models, able to adequately predict human activity in egocentric video sequences. To this end, we introduce a novel three-stream deep neural network architecture combining elements of vision transformers and residual neural networks which are trained using multi-modal data. We evaluate the proposed approach using a challenging, egocentric video dataset and demonstrate its superiority over recent, state-of-the-art research works.

Domain Adaptation for Fear of Heights Classification in a VR Environment Based on EEG and ECG

AbstractThree levels of fear of heights were detected in subjects with different severities of acrophobia, based on the electroencephalographic (EEG) and electrocardiographic (ECG) signals. The study aims to demonstrate the feasibility of a data-fusion-based method for real-time assessment of the fear of heights intensity to integrate into adaptive Virtual Reality Exposure Therapy for acrophobia. The generalization performance of classification tasks on fear states is improved by exploiting both trait-based clustering and Domain Adaptation methods. Participants were gradually exposed to increasing height levels through a Virtual Reality (VR) scenario representing a canyon. The initial severity of fear of heights, the level of distress at each height, and the anxiety level before and after the exposure were assessed through the Acrophobia Questionnaire, the Subjective Unit of Distress, and the State and Trait Anxiety Inventory, respectively. The Simulator Sickness Questionnaire was administered to exclude possible motion sickness interference in the experiment. The EEG and ECG signals were acquired through a 32-channel headset and 1 Lead ECG derivation during the exposure to the eliciting VR scenario. Four classifiers (i.e. Support Vector Machines, Deep Neural Networks, Random Forests, and k-Nearest Neighbors) were adopted in the experimental environment. Preliminary tests were performed in a within-subject experiment, achieving the best classification accuracy of $$87.1 \% \pm 7.8 \%$$ 87.1 % ± 7.8 % with a Deep Neural Network. As the cross-subject approach is concerned, three strategies, namely Domain Adaptation (DA), data fusion (combining EEG with ECG), and participant clustering (based on the acrophobia severity), were evaluated. DA resulted in the most effective strategies by determining an improvement of more than 20 % in classification accuracy. Random Forest performed the best classification accuracy for the severe acrophobia cluster with a mean of $$63.6 \%$$ 63.6 % and a standard deviation of $$ 13.4 \%$$ 13.4 % over three classes by exploiting Stratified Normalization.

A systematic investigation of the transfer of polyphosphate/inorganic silicate flame retardants from epoxy resins to layered glass fiber‐reinforced composites and their post‐furnace flexural properties

AbstractThe systematic transfer of solvent‐free, additive flame retardant (FR) formulations from epoxy resins to glass fiber‐reinforced epoxy composites (GFRECs) through prepregs is difficult. Additionally, obtaining data on their post‐fire mechanics is often challenging. Utilizing melamine polyphosphate (MPP), ammonium polyphosphate (APP), and silane‐coated ammonium polyphosphate (SiAPP) FRs with low‐melting inorganic silicates (InSi) in an 8:2 proportion and 10% loading by weight in a diglycidyl ether of bisphenol A (DGEBA) resin, a systematic investigation of the processing properties, room‐temperature mechanics, and temperature‐based mechanics of the systems was performed. The resin was cured with a dicyandiamide hardener (DICY) and a urone accelerator. The results revealed no substantial impact of these FRs at the current loading on the resin's glass transition temperature or processability. However, the fire residues from cone calorimetry tests of the composites containing FRs were found to be only 15‐20% of the thickness of the resins, implying a suppression of intumescence upon transfer. At room temperature, the decrease in the flexural modulus for the composites containing FRs was negligible. Exposure of the composites in a furnace at 400°C as a preliminary study before ignition tests was shown to cause significant flexural moduli reductions after 2.5 min of exposure and complete delamination after 3 min making further testing unviable. This study emphasizes the need for future research on recovering modes of action upon transfer of FR formulations from resins to composites. Based on the challenges outlined in this investigation, sample adaptation methods for post‐fire analysis will be developed in a future study.Highlights Processibility of resins for prepreg production unaffected by polyphosphate/inorganic silicate flame retardants (FRs). FR formulations had a negligible effect on the mechanics of the composites. 15%–25% increase in the fracture toughness of the DGEBA‐based resin matrix with FRs. Suppression of intumescent behavior in the composites verified quantitatively. Significant reduction in flexural moduli of the composites post‐400°C exposure in a furnace.

Concept-drifts adaptation for machine learning EEG epilepsy seizure prediction.

Seizure prediction remains a challenge, with approximately 30% of patients unresponsive to conventional treatments. Addressing this issue is crucial for improving patients' quality of life, as timely intervention can mitigate the impact of seizures. In this research field, it is critical to identify the preictal interval, the transition from regular brain activity to a seizure. While previous studies have explored various Electroencephalogram (EEG) based methodologies for prediction, few have been clinically applicable. Recent studies have underlined the dynamic nature of EEG data, characterised by data changes with time, known as concept drifts, highlighting the need for automated methods to detect and adapt to these changes. In this study, we investigate the effectiveness of automatic concept drift adaptation methods in seizure prediction. Three patient-specific seizure prediction approaches with a 10-minute prediction horizon are compared: a seizure prediction algorithm incorporating a window adjustment method by optimising performance with Support Vector Machines (Backwards-Landmark Window), a seizure prediction algorithm incorporating a data-batch (seizures) selection method using a logistic regression (Seizure-batch Regression), and a seizure prediction algorithm with a dynamic integration of classifiers (Dynamic Weighted Ensemble). These methods incorporate a retraining process after each seizure and use a combination of univariate linear features and SVM classifiers. The Firing Power was used as a post-processing technique to generate alarms before seizures. These methodologies were compared with a control approach based on the typical machine learning pipeline, considering a group of 37 patients with Temporal Lobe Epilepsy from the EPILEPSIAE database. The best-performing approach (Backwards-Landmark Window) achieved results of 0.75 ± 0.33 for sensitivity and 1.03 ± 1.00 for false positive rate per hour. This new strategy performed above chance for 89% of patients with the surrogate predictor, whereas the control approach only validated 46%.