Training Data Research Articles

Implementasi Data Mining Klasifikasi Gejala Penyakit TB Menggunakan Algoritma Naive Bayes pada Studi Kasus Puskesmas Pegangsaan Dua B

Tuberculosis (TB) is an infectious disease caused by Mycobacterium tuberculosis. The bacteria can enter the lungs, causing the patient to experience shortness of breath and chronic coughing. The infection process in TB progresses through three main stages: primary infection, latent infection, and active infection. During the primary infection stage, the bacteria enter the lungs through inhaled air. If the immune system successfully controls the bacteria, the infection becomes latent, where the bacteria remain inactive and cause no symptoms. However, if the immune system fails, the infection becomes active, and the bacteria attack healthy lung cells, leading to severe symptoms. Tuberculosis remains a major cause of global morbidity and mortality. According to the 2022 WHO report, Indonesia ranks second in the world with an estimated 969,000 TB cases, or 354 per 100,000 people, and a mortality rate of 144,000, or 52 per 100,000 people. This study aims to improve the accuracy of TB diagnosis through the application of the Naive Bayes algorithm, applied to patient data from the Pegangsaan Dua B Community Health Center. The dataset includes 1,527 training data and 1,029 testing data, covering attributes such as patient name, gender, age, identification number, registration date, address, diagnosis status, and diagnosis results. The results indicate that the Naive Bayes algorithm achieved an accuracy of 99.71%, showing its potential for early and accurate TB diagnosis prediction.

Fintech model performance analysis in credit loans: evidence from China city commercial banks

ABSTRACT The analysis of bank loan performance has consistently been a central issue in banking research. With the impact of the COVID-19 pandemic, we have witnessed a significant increase in online loans. Our study employs fintech models to develop a big data credit analysis system, evaluating model efficiency and the effect of ‘data organization’ on predictive performance for city commercial bank loans. Our results show that training data size, timeliness, and inter-bank ‘data sharing’ significantly affect model predictions. We also find that traditional econometric models (Logit and Probit) have the weakest predictive ability. However, ‘most advanced’ deep learning models perform worse than machine learning models with lower algorithmic complexity, such as Random Forest. Besides, ‘sharing training data’ generally increases model efficiency, but there are exceptions for larger city commercial banks. This paper provides ample empirical evidence for the application of the fintech model in bank loan credit. It highlights several characteristics of training data, such as size and timeliness, which greatly affect model efficiencies. We also provide policy recommendations for fintech transition and collaboration among municipal banks.

SeqSeg: Learning Local Segments for Automatic Vascular Model Construction.

Computational modeling of cardiovascular function has become a critical part of diagnosing, treating and understanding cardiovascular disease. Most strategies involve constructing anatomically accurate computer models of cardiovascular structures, which is a multistep, time-consuming process. To improve the model generation process, we herein present SeqSeg (sequential segmentation): a novel deep learning-based automatic tracing and segmentation algorithm for constructing image-based vascular models. SeqSeg leverages local U-Net-based inference to sequentially segment vascular structures from medical image volumes. We tested SeqSeg on CT and MR images of aortic and aortofemoral models and compared the predictions to those of benchmark 2D and 3D global nnU-Net models, which have previously shown excellent accuracy for medical image segmentation. We demonstrate that SeqSeg is able to segment more complete vasculature and is able to generalize to vascular structures not annotated in the training data.

Estimation of Source Range and Location Using Ship-Radiated Noise Measured by Two Vertical Line Arrays with a Feed-Forward Neural Network

Machine learning-based source range estimation is a promising method for enhancing the performance of tracking both the dynamic and static positions of targets in the underwater acoustic environment using extensive training data. This study constructed a machine learning model for source range estimation using ship-radiated noise recorded by two vertical line arrays (VLAs) during the Shallow-water Acoustic Variability Experiment (SAVEX-15), employing the Sample Covariance Matrix (SCM) and the Generalized Cross Correlation (GCC) as input features. A feed-forward neural network (FNN) was used to train the model on the acoustic characteristics of the source at various distances, and the range estimation results indicated that the SCM outperformed the GCC with lower error rates. Additionally, array tilt correction using the array invariant-based method improved range estimation accuracy. The impact of the training data composition corresponding to the bottom depth variation between the source and receivers on range estimation performance was also discussed. Furthermore, the estimated ranges from the two VLA locations were applied to localization using trilateration. Our results confirm that the SCM is the more appropriate feature for the FNN-based source range estimation model compared with the GCC and imply that ocean environment variability should be considered in developing a general-purpose machine learning model for underwater acoustics.

A multimodal approach to support teacher, researcher and AI collaboration in STEM+C learning environments

AbstractRecent advances in generative artificial intelligence (AI) and multimodal learning analytics (MMLA) have allowed for new and creative ways of leveraging AI to support K12 students' collaborative learning in STEM+C domains. To date, there is little evidence of AI methods supporting students' collaboration in complex, open‐ended environments. AI systems are known to underperform humans in (1) interpreting students' emotions in learning contexts, (2) grasping the nuances of social interactions and (3) understanding domain‐specific information that was not well‐represented in the training data. As such, combined human and AI (ie, hybrid) approaches are needed to overcome the current limitations of AI systems. In this paper, we take a first step towards investigating how a human‐AI collaboration between teachers and researchers using an AI‐generated multimodal timeline can guide and support teachers' feedback while addressing students' STEM+C difficulties as they work collaboratively to build computational models and solve problems. In doing so, we present a framework characterizing the human component of our human‐AI partnership as a collaboration between teachers and researchers. To evaluate our approach, we present our timeline to a high school teacher and discuss the key insights gleaned from our discussions. Our case study analysis reveals the effectiveness of an iterative approach to using human‐AI collaboration to address students' STEM+C challenges: the teacher can use the AI‐generated timeline to guide formative feedback for students, and the researchers can leverage the teacher's feedback to help improve the multimodal timeline. Additionally, we characterize our findings with respect to two events of interest to the teacher: (1) when the students cross a difficulty threshold, and (2) the point of intervention, that is, when the teacher (or system) should intervene to provide effective feedback. It is important to note that the teacher explained that there should be a lag between (1) and (2) to give students a chance to resolve their own difficulties. Typically, such a lag is not implemented in computer‐based learning environments that provide feedback. Practitioner notesWhat is already known about this topic Collaborative, open‐ended learning environments enhance students' STEM+C conceptual understanding and practice, but they introduce additional complexities when students learn concepts spanning multiple domains. Recent advances in generative AI and MMLA allow for integrating multiple datastreams to derive holistic views of students' states, which can support more informed feedback mechanisms to address students' difficulties in complex STEM+C environments. Hybrid human‐AI approaches can help address collaborating students' STEM+C difficulties by combining the domain knowledge, emotional intelligence and social awareness of human experts with the general knowledge and efficiency of AI. What this paper adds We extend a previous human‐AI collaboration framework using a hybrid intelligence approach to characterize the human component of the partnership as a researcher‐teacher partnership and present our approach as a teacher‐researcher‐AI collaboration. We adapt an AI‐generated multimodal timeline to actualize our human‐AI collaboration by pairing the timeline with videos of students encountering difficulties, engaging in active discussions with a high school teacher while watching the videos to discern the timeline's utility in the classroom. From our discussions with the teacher, we define two types of inflection points to address students' STEM+C difficulties—the difficulty threshold and the intervention point—and discuss how the feedback latency interval separating them can inform educator interventions. We discuss two ways in which our teacher‐researcher‐AI collaboration can help teachers support students encountering STEM+C difficulties: (1) teachers using the multimodal timeline to guide feedback for students, and (2) researchers using teachers' input to iteratively refine the multimodal timeline. Implications for practice and/or policy Our case study suggests that timeline gaps (ie, disengaged behaviour identified by off‐screen students, pauses in discourse and lulls in environment actions) are particularly important for identifying inflection points and formulating formative feedback. Human‐AI collaboration exists on a dynamic spectrum and requires varying degrees of human control and AI automation depending on the context of the learning task and students' work in the environment. Our analysis of this human‐AI collaboration using a multimodal timeline can be extended in the future to support students and teachers in additional ways, for example, designing pedagogical agents that interact directly with students, developing intervention and reflection tools for teachers, helping teachers craft daily lesson plans and aiding teachers and administrators in designing curricula.

AutoPeptideML: A study on how to build more trustworthy peptide bioactivity predictors.

Automated machine learning (AutoML) solutions can bridge the gap between new computational advances and their real-world applications by enabling experimental scientists to build their own custom models. We examine different steps in the development life-cycle of peptide bioactivity binary predictors and identify key steps where automation can not only result in a more accessible method, but also more robust and interpretable evaluation leading to more trustworthy models. We present a new automated method for drawing negative peptides that achieves better balance between specificity and generalisation than current alternatives. We study the effect of homology-based partitioning for generating the training and testing data subsets and demonstrate that model performance is overestimated when no such homology correction is used, which indicates that prior studies may have overestimated their performance when applied to new peptide sequences. We also conduct a systematic analysis of different protein language models as peptide representation methods and find that they can serve as better descriptors than a naive alternative, but that there is no significant difference across models with different sizes or algorithms. Finally, we demonstrate that an ensemble of optimised traditional machine learning algorithms can compete with more complex neural network models, while being more computationally efficient. We integrate these findings into AutoPeptideML, an easy-to-use AutoML tool to allow researchers without a computational background to build new predictive models for peptide bioactivity in a matter of minutes. Source code, documentation, and data are available at https://github.com/IBM/AutoPeptideML and a dedicated web-server at http://peptide.ucd.ie/AutoPeptideML. Raul.fernandezdiaz@ucdconnect.ie or denis.shields@ucd.ie. Supplementary Information can be accessed from Zenodo: Https://zenodo.org/records/13363975.

Transient NOx emission modeling of a hydrogen-diesel engine using hybrid machine learning methods

One promising approach to reduce carbon foot print of internal combustion engines (ICEs) is using alternative fuels like hydrogen, particularly by converting medium and heavy-duty diesel engines to dual-fuel hydrogen-diesel engines. To minimize elevated NOx emissions from hydrogen-fueled engine, fast and accurate emission models are essential for engine model-based control and for engine calibration and optimization using hardware-in-the-loop (HIL) setups. In this study, a fast-response NOx emissions sensor is used to measure the transient NOx emissions from a dual-fuel hydrogen-diesel engine. Subsequently, steady-state models (SSMs), quasi steady-state models (QSSMs), and transient sequential models (TSMs) in the form of black-box (BB) and gray-box (GB) models are developed for transient NOx emissions prediction. GB models utilize both information from a one dimensional (1D) physical engine model and experimental data for training, while BB models only use experimental data. SSMs are optimized artificial neural networks (ANNs) trained using steady-state data, QSSMs are optimized ANNs trained using transient data, and TSMs are time-series networks trained using transient data. Long short-term memory (LSTM) and gated recurrent unit (GRU) networks are used as the time-series deep learning networks. The results showed that the 1D physical model has the poorest performance with successive model performance improvement from SSM to QSSM and from QSSM to TSM. The developed BB TSM model in this study can predict transient NOx emissions with an R2 value greater than 0.96 at 89,000 predictions per second which makes this model suitable for real-time engine model-based control where computational efficiency is crucial. The developed GB TSM model can predict transient NOx emissions with an R2 value greater than 0.97 but it is computationally more expensive. The extra accuracy of the GB TSM models makes them the best choice for HIL setups where more computational power is available, and accuracy is more crucial.

Artificial neural network-based control of powered knee exoskeletons for lifting tasks: design and experimental validation

Abstract This study introduces a hybrid model that utilizes a model-based optimization method to generate training data and an artificial neural network (ANN)-based learning method to offer real-time exoskeleton support in lifting activities. For the model-based optimization method, the torque of the knee exoskeleton and the optimal lifting motion are predicted utilizing a two-dimensional (2D) human–exoskeleton model. The control points for exoskeleton motor current profiles and human joint angle profiles from cubic B-spline interpolation represent the design variables. Minimizing the square of the normalized human joint torque is considered as the cost function. Subsequently, the lifting optimization problem is tackled using a sequential quadratic programming (SQP) algorithm in sparse nonlinear optimizer (SNOPT). For the learning-based approach, the learning-based control model is trained using the general regression neural network (GRNN). The anthropometric parameters of the human subjects and lifting boundary postures are used as input parameters, while the control points for exoskeleton torque are treated as output parameters. Once trained, the learning-based control model can provide exoskeleton assistive torque in real time for lifting tasks. Two test subjects’ joint angles and ground reaction forces (GRFs) comparisons are presented between the experimental and simulation results. Furthermore, the utilization of exoskeletons significantly reduces activations of the four knee extensor and flexor muscles compared to lifting without the exoskeletons for both subjects. Overall, the learning-based control method can generate assistive torque profiles in real time and faster than the model-based optimal control approach.

Latent Diffusion Models to Enhance the Performance of Visual Defect Segmentation Networks in Steel Surface Inspection

This paper explores the use of state-of-the-art latent diffusion models, specifically stable diffusion, to generate synthetic images for improving the robustness of visual defect segmentation in manufacturing components. Given the scarcity and imbalance of real-world defect data, synthetic data generation offers a promising solution for training deep learning models. We fine-tuned stable diffusion using the LoRA technique on the NEU-seg dataset and evaluated the impact of different ratios of synthetic to real images on the training set of DeepLabV3+ and FPN segmentation models. Our results demonstrated a significant improvement in mean Intersection over Union (mIoU) when the training dataset was augmented with synthetic images. This study highlights the potential of diffusion models for enhancing the quality and diversity of training data in industrial defect detection, leading to more accurate and reliable segmentation results. The proposed approach achieved improvements of 5.95% and 6.85% in mIoU of defect segmentation on each model over the original dataset.

TPLAD: Template-Parsed Log Anomaly Detection for Electrical Database Systems

In recent years, with the increasing complexity of software systems, logs have become crucial for system maintenance. Log-based anomaly detection plays a vital role in automatically detecting system anomalies through log analysis. However, current log-based anomaly detection approaches encounter significant practical challenges. Supervised methods often require a large amount of manually labeled training data, which can be time-consuming and costly to obtain. On the other hand, unsupervised and semi-supervised approaches may suffer from subpar performance, as they do not leverage historical anomalies to improve their detection capabilities. These challenges underscore the necessity for the development of more efficient and effective log-based anomaly detection methods. Database anomaly access detection is critical for ensuring the stability and security of database systems. We present a survey of existing log anomaly detection models and propose a novel approach, Template-Parsed Log Anomaly Detection (TPLAD) model, for automated anomaly detection in massive database log files. The proposed model combines the original log template with template parsing using code and text semantic representation. Experimental results demonstrate the effectiveness of the proposed approach in detecting abnormal database access patterns, including runtime errors, unauthorized access, and data leaks. The findings indicate that TPLAD model shows promise in enhancing database security and stability in business systems.

Artificial Intelligence Tools in Pediatric Urology: A Comprehensive Review of Recent Advances

Artificial intelligence (AI) is providing novel answers to long-standing clinical problems, and it is quickly changing pediatric urology. This thorough analysis focuses on current developments in AI technologies that improve pediatric urology diagnosis, treatment planning, and surgery results. Deep learning algorithms help detect problems with previously unheard-of precision in disorders including hydronephrosis, pyeloplasty, and vesicoureteral reflux, where AI-powered prediction models have demonstrated promising outcomes in boosting diagnostic accuracy. AI-enhanced image processing methods have significantly improved the quality and interpretation of medical images. Examples of these methods are deep-learning-based segmentation and contrast limited adaptive histogram equalization (CLAHE). These methods guarantee higher precision in the identification and classification of pediatric urological disorders, and AI-driven ground truth construction approaches aid in the standardization of and improvement in training data, resulting in more resilient and consistent segmentation models. AI is being used for surgical support as well. AI-assisted navigation devices help with difficult operations like pyeloplasty by decreasing complications and increasing surgical accuracy. AI also helps with long-term patient monitoring, predictive analytics, and customized treatment strategies, all of which improve results for younger patients. However, there are practical, ethical, and legal issues with AI integration in pediatric urology that need to be carefully navigated. To close knowledge gaps, more investigation is required, especially in the areas of AI-driven surgical methods and standardized ground truth datasets for pediatric radiologic image segmentation. In the end, AI has the potential to completely transform pediatric urology by enhancing patient care, increasing the effectiveness of treatments, and spurring more advancements in this exciting area.

In Silico Insights: QSAR Modeling of TBK1 Kinase Inhibitors for Enhanced Drug Discovery.

TBK1, or TANK-binding kinase 1, is an enzyme that functions as a serine/threonine protein kinase. It plays a crucial role in various cellular processes, including the innate immune response to viruses, cell proliferation, apoptosis, autophagy, and antitumor immunity. Dysregulation of TBK1 activity can lead to autoimmune diseases, neurodegenerative disorders, and cancer. Due to its central role in these critical pathways, TBK1 is a significant focus of research for therapeutic drug development. In this paper, we explore data from the CAS Content Collection regarding TBK1 and its implication in a large assortment of diseases and disorders. With the demand for developing efficient TBK1 inhibitors being outlined, we focus on utilizing a machine learning approach for developing predictive models for TBK1 inhibition, derived from the fragment-functional analysis descriptors. Using the extensive CAS Content Collection, we assembled a training set of TBK1 inhibitors with experimentally measured IC50 values. We explored several machine learning techniques combined with various molecular descriptors to derive and select the best TBK1 inhibitor QSAR models. Certain significant structural alerts that potentially contribute to inhibition of TBK1 are outlined and discussed. The merit of the article stems from identifying the most adequate TBK1 QSAR models and subsequent successful development of advanced positive training data to facilitate and enhance drug discovery for an important therapeutic target such as TBK1 inhibitors, based on an extensive, wide-ranging set of scientific information provided by the CAS Content Collection.

Robust optical picometrology through data diversity

Topologically structured light contains deeply subwavelength features, such as phase singularities, and the scattering of such light can therefore be sensitive to the geometry or movement of scattering objects at such scales. Indeed, it has been shown recently that single-shot optical measurements can yield positional precision better than 100 pm (less than one five-thousandth of the wavelength λ) via a deep-learning-enabled analysis of scattering patterns. Measurement performance, and the extent to which it can be sustained, are constrained by the quality and depth of neural network training data and the stability of the experimental apparatus. Here, we show that a neural network can be trained through exposure to an extended envelope of instrumental/ambient noise conditions to robustly quantify picometric displacements of a target against orders-of-magnitude larger background fluctuations, to maintain precision and accuracy of 100–150 pm in optical measurements (at λ = 488 nm) of nanowire positional change. This capability opens up a range of application opportunities, for example in the optical study of nanostructural dynamics, stiction, material fatigue, and phase transitions.

Multi-Objective Design of DNA-Stabilized Nanoclusters Using Variational Autoencoders With Automatic Feature Extraction.

DNA-stabilized silver nanoclusters (AgN-DNAs) have sequence-tuned compositions and fluorescence colors. High-throughput experiments together with supervised machine learning models have recently enabled design of DNA templates that select for AgN-DNA properties, including near-infrared (NIR) emission that holds promise for deep tissue bioimaging. However, these existing models do not enable simultaneous selection of multiple AgN-DNA properties, and require significant expert input for feature engineering and class definitions. This work presents a model for multiobjective, continuous-property design of AgN-DNAs with automatic feature extraction, based on variational autoencoders (VAEs). This model is generative, i.e., it learns both the forward mapping from DNA sequence to AgN-DNA properties and the inverse mapping from properties to sequence, and is trained on an experimental data set of DNA sequences paired with AgN-DNA fluorescence properties. Experimental testing shows that the model enables effective design of AgN-DNA emission, including bright NIR AgN-DNAs with 4-fold greater abundance compared to training data. In addition, Shapley analysis is employed to discern learned nucleobase patterns that correspond to fluorescence color and brightness. This generative model can be adapted for a range of biomolecular systems with sequence-dependent properties, enabling precise design of emerging biomolecular nanomaterials.

Enhanced 2D Hand Pose Estimation for Gloved Medical Applications: A Preliminary Model

(1) Background: As digital health technology evolves, the role of accurate medical-gloved hand tracking is becoming more important for the assessment and training of practitioners to reduce procedural errors in clinical settings. (2) Method: This study utilized computer vision for hand pose estimation to model skeletal hand movements during in situ aseptic drug compounding procedures. High-definition video cameras recorded hand movements while practitioners wore medical gloves of different colors. Hand poses were manually annotated, and machine learning models were developed and trained using the DeepLabCut interface via an 80/20 training/testing split. (3) Results: The developed model achieved an average root mean square error (RMSE) of 5.89 pixels across the training data set and 10.06 pixels across the test set. When excluding keypoints with a confidence value below 60%, the test set RMSE improved to 7.48 pixels, reflecting high accuracy in hand pose tracking. (4) Conclusions: The developed hand pose estimation model effectively tracks hand movements across both controlled and in situ drug compounding contexts, offering a first-of-its-kind medical glove hand tracking method. This model holds potential for enhancing clinical training and ensuring procedural safety, particularly in tasks requiring high precision such as drug compounding.

Debiasing large language models: research opportunities*

ABSTRACT Large language models (LLMs) are powerful decision-making tools widely adopted in healthcare, finance, and transportation. Embracing the opportunities and innovations of LLMs is inevitable. However, LLMs inherit stereotypes, misrepresentations, discrimination, and societies' biases from various sources–including training data, algorithm design, and user interactions–resulting in concerns about equality, diversity, and fairness. The bias problem has triggered increased research towards defining, detecting and quantifying bias and developing debiasing techniques. The predominant focus in tackling the bias problem is skewed towards resource-rich regions such as the US and Europe, resulting in a scarcity of research in other societies. As a small country with a unique history, culture and social composition, there is an opportunity for Aotearoa New Zealand's (NZ) research community to address this inadequacy. This paper presents an experimental evaluation of existing bias metrics and debiasing techniques in the NZ context. Research gaps derived from the study and a literature review are outlined, current and ongoing research in this space are discussed, and the suggested scope of research opportunities for NZ are presented.

Prediction of the Cu oxidation state from EELS and XAS spectra using supervised machine learning

Electron energy loss spectroscopy (EELS) and X-ray absorption spectroscopy (XAS) provide detailed information about bonding, distributions and locations of atoms, and their coordination numbers and oxidation states. However, analysis of XAS/EELS data often relies on matching an unknown experimental sample to a series of simulated or experimental standard samples. This limits analysis throughput and the ability to extract quantitative information from a sample. In this work, we have trained a random forest model capable of predicting the oxidation state of copper based on its L-edge spectrum. Our model attains an R2 score of 0.85 and a root mean square error of 0.24 on simulated data. It has also successfully predicted experimental L-edge EELS spectra taken in this work and XAS spectra extracted from the literature. We further demonstrate the utility of this model by predicting simulated and experimental spectra of mixed valence samples generated by this work. This model can be integrated into a real-time EELS/XAS analysis pipeline on mixtures of copper-containing materials of unknown composition and oxidation state. By expanding the training data, this methodology can be extended to data-driven spectral analysis of a broad range of materials.

Retraction Note: Light image enhancement and virtual reality application in automatic generation of basketball game scenes and training data simulation

Retraction Note: Light image enhancement and virtual reality application in automatic generation of basketball game scenes and training data simulation

A Survey of Mix-based Data Augmentation: Taxonomy, Methods, Applications, and Explainability

Data augmentation (DA) is indispensable in modern machine learning and deep neural networks. The basic idea of DA is to construct new training data to improve the model’s generalization by adding slightly disturbed versions of existing data or synthesizing new data. This survey comprehensively reviews a crucial subset of DA techniques, namely Mix-based Data Augmentation (MixDA), which generates novel samples by combining multiple examples. In contrast to traditional DA approaches that operate on single samples or entire datasets, MixDA stands out due to its effectiveness, simplicity, computational efficiency, theoretical foundation, and broad applicability. We begin by introducing a novel taxonomy that categorizes MixDA into Mixup-based, Cutmix-based, and mixture approaches based on a hierarchical perspective of the data mixing operation. Subsequently, we provide an in-depth review of various MixDA techniques, focusing on their underlying motivations. Owing to its versatility, MixDA has penetrated a wide range of applications, which we also thoroughly investigate in this survey. Moreover, we delve into the underlying mechanisms of MixDA’s effectiveness by examining its impact on model generalization and calibration while providing insights into the model’s behavior by analyzing the inherent properties of MixDA. Finally, we recapitulate the critical findings and fundamental challenges of current MixDA studies while outlining the potential directions for future works. Different from previous related surveys that focus on DA approaches in specific domains (e.g., computer vision and natural language processing) or only review a limited subset of MixDA studies, we are the first to provide a systematical survey of MixDA, covering its taxonomy, methodology, application, and explainability. Furthermore, we provide promising directions for researchers interested in this exciting area.

Automated Detection of Hillforts in Remote Sensing Imagery With Deep Multimodal Segmentation

ABSTRACTRecent advancements in remote sensing and artificial intelligence can potentially revolutionize the automated detection of archaeological sites. However, the challenging task of interpreting remote sensing imagery combined with the intricate shapes of archaeological sites can hinder the performance of computer vision systems. This work presents a computer vision system trained for efficient hillfort detection in remote sensing imagery. Equipped with an adapted multimodal semantic segmentation model, the system integrates LiDAR‐derived LRM images and aerial orthoimages for feature fusion, generating a binary mask pinpointing detected hillforts. Post‐processing includes margin and area filters to remove edge inferences and smaller anomalies. The resulting inferences are subjected to hard positive and negative mining, where expert archaeologists classify them to populate the training data with new samples for retraining the segmentation model. As the computer vision system is far more likely to encounter background images during its search, the training data are intentionally biased towards negative examples. This approach aims to reduce the number of false positives, typically seen when applying machine learning solutions to remote sensing imagery. Northwest Iberia experiments witnessed a drastic reduction in false positives, from 5678 to 40 after a single hard positive and negative mining iteration, yielding a 99.3% reduction, with a resulting F1 score of 66%. In England experiments, the system achieved a 59% F1 score when fine‐tuned and deployed countrywide. Its scalability to diverse archaeological sites is demonstrated by successfully detecting hillforts and other types of enclosures despite their typical complex and varied shapes. Future work will explore archaeological predictive modelling to identify regions with higher archaeological potential to focus the search, addressing processing time challenges.