XLXiaowen Liu

Xiaowen Liu

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Xiaowen Liu has a total of 18 co-authors. Here are the top co-authors by number of shared publications:

  • Abdulkareem Alzahrani

    Computer Science Department, Faculty of Computing and Information, Al-Baha University, Al-Baha, Saudi Arabia

    3 shared publications
  • Anas Bilal

    College of Information Science and Technology Hainan Normal University Haikou China

    3 shared publications
  • Khalid Almohammadi

    Computer Science Department, Applied College, University of Tabuk, Tabuk, Saudi Arabia

    2 shared publications
  • Talha Imtiaz Baig

    The Clinical Hospital of Chengdu Brain Science Institute, MOE Key Laboratory for Neuroinformation, School of Life Science and Technology, University of Electronic Science and Technology of China, No.2006, Xiyuan Avenue, West Hi-Tech Zone, Chengdu, Sichuan 611731, China; Key Laboratory of Psychosomatic Medicine, Chinese Academy of Medical Sciences, Chengdu 610072, China

    2 shared publications
  • Anas Bilal

    Department of Information Science and Technology, Hainan Normal University , No. 99 Long Kun South Road, Haikou 571158

    1 shared publication
  • Azhar Imran

    Department of Creative Technologies, Air University, Islamabad, Pakistan

    1 shared publication

Latest Publications

  • Open Access

    An LLM-Based Multi-Path Question Answering System with XGBoost Routing and Threshold-Based Refusal

    by Bo Dai, Caiyun Li, Yiyun Cao, Jie Ling, Xiaowen Liu

    In conventional question answering systems, general-purpose large language models (LLMs), despite their strong capabilities in language understanding and generation, exhibit notable limitations in scenarios with stringent factuality requirements. Their outputs often lack explicit evidential grounding, making them prone to hallucinations and inconsistent responses. Moreover, LLMs do not inherently guarantee determinism when performing operations over structured data—such as aggregation, conditional filtering, and cross-field constraints—thereby undermining result consistency and reliability. To address these issues, we propose an internal-data-first framework for controllable question answering in high-risk scenarios. The framework categorizes knowledge sources into unstructured documents and structured data, enabling evidence-constrained generation via retrieval-augmented generation (RAG) and database-backed, verifiable query execution via restricted, read-only structured queries. In addition, an UNK branch is introduced as a safe degradation mechanism that triggers refusal when inputs lack sufficient evidence, exceed system capability boundaries, or fail to meet confidence requirements, thereby suppressing hallucinations and unauthorized generation. To enable controlled selection among the two execution pathways (RAG/SQL) and the safety degradation branch (UNK) at the system level, we design a learned router based on XGBoost with confidence-thresholded selective prediction, which preferentially activates UNK refusals for low-confidence or out-of-distribution inputs. We validate the proposed framework using a graduate admissions consultation system as an exemplar application, constructing both a document knowledge base and structured score tables, and conducting controlled comparisons across multiple system variants with multi-metric evaluations. Experimental results indicate that, under the current controlled evaluation setting, the proposed framework exhibits relatively stable behavior under complex query formulations and demonstrates practical engineering potential in high-risk vertical-domain question answering scenarios.


  • Open Access

    Fuzzy deep learning architecture for cucumber plant disease detection and classification

    by Anas Bilal, Junaid Ali Khan, Abdulkareem Alzahrani, Khalid Almohammadi, Maha Alamri, Xiaowen Liu


  • Open Access

    Research on the Application of Single-Parent Genetic Algorithm Improved by Sine Chaotic Mapping in Parent–Child Travel Path Optimization

    by Zhi-Heng Wang, Xiao-Wen Liu

    This paper proposes a method for recommending parent–child travel destinations and planning travel routes tailored to children of different ages. The method inputs basic information about the attractions (such as ticket prices, geographical locations, opening hours, etc.) into the system database and intelligently recommends suitable attractions based on user-provided data, including the children’s age, travel time, and trip theme. The paper transforms the route planning problem into a Traveling Salesman Problem (TSP) to optimize the travel route further. It presents an improved single-parent genetic algorithm based on sine chaos mapping (SCM-SPGA) to solve and optimize the shortest path for parent–child trips. Experimental results demonstrate that this algorithm has significant advantages in path planning accuracy and efficiency. The method is applied to a tourism dataset of Hainan, providing more personalized and age-appropriate attraction recommendations for tourists planning a parent–child trip to Hainan and optimizing the travel route. The research shows that the proposed method can effectively meet the personalized needs of parent–child travelers, significantly improving the overall travel experience by offering more tailored, efficient, and enjoyable trip-planning solutions.


  • Open Access

    2OM-Pred: prediction of 2-O-methylation sites in ribonucleic acid using diverse classifiers

    by Anas Bilal, Muhammad Taseer Suleman, Khalid Almohammadi, Abdulkareem Alzahrani, Xiaowen Liu

    2-O-methylation (2OM) is a vital post-transcriptional modification which is formed by a functional group through the attachment of a methyl (–CH3) group to the second position of an aromatic ring hydroxyl group (–OH). It plays an active part in RNA physical configuration stability and the way different RNA molecules interrelate. Further, this modification plays a pivotal role in changing the epigenetic regulation of cellular processes. Previous approaches like mass spectrometry could not fully enhance the identification of RNA-modified sites. Sequence data were useful in the development of measures that meant the use of computationally intelligent system to identify 2OM sites quickly. This research proposed a new novel method of feature extraction and generation from the available sequences, and the feature dimensionality reduction has been done through the incorporation of statistical moments. The final feature vectors were developed and used to train prediction models. The assessment of prediction models was carried out through independent set tests and k-fold cross-validation. Through rigorous testing, the bagging ensemble model outperformed and revealed optimal accuracy scores. A publicly accessible web-based application has been developed which can be accessed via https://2om-pred-webapp.streamlit.app/.


  • Open Access

    DeepSVDNet: A Deep Learning-Based Approach for Detecting and Classifying Vision-Threatening Diabetic Retinopathy in Retinal Fundus Images

    by Anas Bilal, Azhar Imran, Talha Imtiaz Baig, Xiaowen Liu, Haixia Long, Abdulkareem Alzahrani, Muhammad Shafiq

    Artificial Intelligence (AI) is being increasingly used for diagnosing Vision-Threatening Diabetic Retinopathy (VTDR), which is a leading cause of visual impairment and blindness worldwide. However, previous automated VTDR detection methods have mainly relied on manual feature extraction and classification, leading to errors. This paper proposes a novel VTDR detection and classification model that combines different models through majority voting. Our proposed methodology involves preprocessing, data augmentation, feature extraction, and classification stages. We use a hybrid convolutional neural network-singular value decomposition (CNN-SVD) model for feature extraction and selection and an improved SVM-RBF with a Decision Tree (DT) and K-Nearest Neighbor (KNN) for classification. We tested our model on the IDRiD dataset and achieved an accuracy of 98.06%, a sensitivity of 83.67%, and a specificity of 100% for DR detection and evaluation tests, respectively. Our proposed approach outperforms baseline techniques and provides a more robust and accurate method for VTDR detection.


  • Open Access

    EdgeSVDNet: 5G-Enabled Detection and Classification of Vision-Threatening Diabetic Retinopathy in Retinal Fundus Images

    by Anas Bilal, Xiaowen Liu, Talha Imtiaz Baig, Haixia Long, Muhammad Shafiq

    The rise of vision-threatening diabetic retinopathy (VTDR) underscores the imperative for advanced and efficient early detection mechanisms. With the integration of the Internet of Things (IoT) and 5G technologies, there is transformative potential for VTDR diagnosis, facilitating real-time processing of the burgeoning volume of fundus images (FIs). Combined with artificial intelligence (AI), this offers a robust platform for managing vast healthcare datasets and achieving unparalleled disease detection precision. Our study introduces a novel AI-driven VTDR detection framework that integrates multiple models through majority voting. This comprehensive approach encompasses pre-processing, data augmentation, feature extraction using a hybrid convolutional neural network-singular value decomposition (CNN-SVD) model, and classification through an enhanced SVM-RBF combined with a decision tree (DT) and K-nearest neighbor (KNN). Validated on the IDRiD dataset, our model boasts an accuracy of 99.89%, a sensitivity of 84.40%, and a specificity of 100%, marking a significant improvement over the traditional method. The convergence of the IoT, 5G, and AI technologies herald a transformative era in healthcare, ensuring timely and accurate VTDR diagnoses, especially in geographically underserved regions.