A5013904947- Complex Network Analysis Techniques
- Lung Cancer Treatments and Mutations
- scientometrics and bibliometrics research
- Radiomics and Machine Learning in Medical Imaging
- Expert finding and Q&A systems
- Topic Modeling
- Lung Cancer Diagnosis and Treatment
- Bioinformatics and Genomic Networks
- Web visibility and informetrics
- Machine Learning in Healthcare
- Artificial Intelligence in Healthcare
- Gene expression and cancer classification
- Computational Drug Discovery Methods
- Privacy-Preserving Technologies in Data
- Scientific Computing and Data Management
- Data Quality and Management
- Artificial Intelligence in Healthcare and Education
- Ferroptosis and cancer prognosis
- AI in cancer detection
- Anomaly Detection Techniques and Applications
- Statistical Methods in Clinical Trials
- Big Data and Business Intelligence
- Antibiotic Use and Resistance
- Customer churn and segmentation
- Domain Adaptation and Few-Shot Learning
Ollscoil na Gaillimhe – University of Galway
2016-2025
Asian Institute of Technology
2012
Despite the growing popularity of machine learning models in cyber‐security applications (e.g., an intrusion detection system (IDS)), most these are perceived as a black‐box. The eXplainable Artificial Intelligence (XAI) has become increasingly important to interpret enhance trust management by allowing human experts understand underlying data evidence and causal reasoning. According IDS, critical role is impact malicious detect any system. previous studies focused more on accuracy various...
Identifying the unintended effects of drugs (side effects) is a very important issue in pharmacological studies. The laboratory verification associations between and side requires costly, time-intensive research. Thus, an approach to predicting drug based on known effects, using computational model, highly desirable. To provide such we used openly available data resources model as bipartite graph. drug-drug network constructed word2vec where edges represent semantic similarity them. We...
In real-world machine learning applications, unlabeled training data are readily available, but labeled expensive and hard to obtain. Therefore, semi-supervised algorithms have gathered much attention. Previous studies in this area mainly focused on a classification problem, whereas regression has received less paper, we proposed novel algorithm using heat diffusion with boundary-condition that guarantees closed-form solution. Experiments from artificial real datasets business, biomedical,...
PURPOSE Stratifying patients with cancer according to risk of relapse can personalize their care. In this work, we provide an answer the following research question: How use machine learning estimate probability in early-stage non–small-cell lung (NSCLC)? MATERIALS AND METHODS For predicting 1,387 (I-II) NSCLC from Spanish Lung Cancer Group data (average age 65.7 years, female 24.8%, male 75.2%), train tabular and graph models. We generate automatic explanations for predictions such models...
Amidst prevailing healthcare challenges, a dynamic solution emerges, fusing knowledge graph technology, clinical trials optimization, dataspace integration, and AI innovation. This unified approach tackles issues like limited patient insights, suboptimal trial designs, imprecise treatments. By interlinking diverse data through graphs, this method illuminates disease trends, therapeutic efficacies, prognoses. techniques, especially machine learning, contribute predictive power by unveiling...
The study addresses customer churn, a major issue in service-oriented sectors like telecommunications, where it refers to the discontinuation of subscriptions. research emphasizes importance recognizing satisfaction for retaining clients, focusing specifically on early churn prediction as key strategy. Previous approaches mainly used generalized classification techniques but often neglected aspect interpretability, vital decision-making. This introduces explainer models address this gap,...
Determining the association between tumor sample and gene is demanding because it requires a high cost for conducting genetic experiments. Thus, discovered further clinical verification validation. This entire mechanism time-consuming expensive. Due to this issue, predicting samples genes remain challenge in biomedicine.Here we present, computational model based on heat diffusion algorithm which can predict genes. We proposed 2-layered graph. In first layer, constructed graph of where these...
Background: Stratifying cancer patients according to risk of relapse can personalize their care. In this work, we provide an answer the following research question: How utilize machine learning estimate probability in early-stage non-small-cell lung patients? Methods: For predicting 1,387 (I-II), (NSCLC) from Spanish Lung Cancer Group data (65.7 average age, 24.8% females, 75.2% males) train tabular and graph models. We generate automatic explanations for predictions such models trained on...
The recurrence of low-stage lung cancer poses a challenge due to its unpredictable nature and diverse patient responses treatments. Personalized care outcomes heavily rely on early relapse identification, yet current predictive models, despite their potential, lack comprehensive genetic data. This inadequacy fuels our research focus—integrating specific information, such as pathway scores, into clinical Our aim is refine machine learning models for more precise prediction in early-stage...
The study addresses customer churn, a major issue in service-oriented sectors like telecommunications, where it refers to the discontinuation of subscriptions. research emphasizes importance recognizing satisfaction for retaining clients, focusing specifically on early churn prediction as key strategy. Previous approaches mainly used generalized classification techniques but often neglected aspect interpretability, vital decision-making. This introduces explainer models address this gap,...
The realization that scholarly publications are discussed and have influence on discourse outside scientific academic domains has given rise to area of scientometrics called alternative metrics or "altmetrics". Furthermore, researchers in this field tend focus primarily measuring activity social media platforms such as Twitter, however these count-based vulnerable gaming because they lack concrete justification reference the primary source. In collaboration with Elsevier, we extend...
The realization that scholarly publications are discussed and have influence on discourse outside scientific academic domains has given rise to area of scientometrics called alternative metrics or altmetrics. Furthermore, researchers in this field tend focus primarily measuring activity social media platforms such as Twitter, however these count-based vulnerable gaming because they lack concrete justification reference the primary source. In collaboration with Elsevier, we extend...
For online social networks, demographic analysis is absolutely essential for improving their services in many ways. It instrumental understanding different audiences, members and competitors. As well as that, it pivotal designing effective personalization contextualization strategies, especially displaying creating better content. There is, this reason, a great bulk of research into how variables are characterized they impact platforms such Facebook Twitter. But surprisingly, only handful...
Lung cancer exhibits unpredictable recurrence in low-stage tumors and variable responses to different therapeutic interventions. Predicting relapse early-stage lung can facilitate precision medicine improve patient survivability. While existing machine learning models rely on clinical data, incorporating genomic information could enhance their efficiency. This study aims impute integrate specific types of data with the accuracy for predicting early-stage, non-small cell patients.The utilized...
Semi-Supervised Learning (SSL)is an approach to machine learning that makes use of unlabeled data for training with a small amount labeled data. In the context molecular biology and pharmacology, one can take advantage For instance, identify drugs targets where few genes are known be associated specific target considered as Labeling requires laboratory verification validation. This process is usually very time consuming expensive. Thus, it useful estimate functional role from using...
The growth in the alternative digital publishing is widening breadth of scholarly impact beyond conventional bibliometric community. Thus, research becoming more reachable both inside and outside academic institutions are found to be shared, downloaded discussed social media. In this study, we linked scientific articles mainstream news, weblogs Stack Overflow citation database peer-reviewed literature called Scopus. We then explored how standard graph-based influence metrics can used measure...