Gastric cancer and Colon adenocarcinoma represent widespread and challenging malignancies with high mortality rates and complex treatment landscapes. In response to the critical need for accurate prognosis in cancer patients, the medical community has embraced the 5-year survival rate as a vital metric for estimating patient outcomes. This study introduces a pioneering approach to enhance survival prediction models for gastric and Colon adenocarcinoma patients. Leveraging advanced image analysis techniques, we sliced whole slide images (WSI) of these cancers, extracting comprehensive features to capture nuanced tumor characteristics. Subsequently, we constructed patient-level graphs, encapsulating intricate spatial relationships within tumor tissues. These graphs served as inputs for a sophisticated 4-layer graph convolutional neural network (GCN), designed to exploit the inherent connectivity of the data for comprehensive analysis and prediction. By integrating patients’ total survival time and survival status, we computed C-index values for gastric cancer and Colon adenocarcinoma, yielding 0.57 and 0.64, respectively. Significantly surpassing previous convolutional neural network models, these results underscore the efficacy of our approach in accurately predicting patient survival outcomes. This research holds profound implications for both the medical and AI communities, offering insights into cancer biology and progression while advancing personalized treatment strategies. Ultimately, our study represents a significant stride in leveraging AI-driven methodologies to revolutionize cancer prognosis and improve patient outcomes on a global scale.

To address the issues of limited samples, time-consuming feature design, and low accuracy in detection and classification of breast cancer pathological images, a breast cancer image classification model algorithm combining deep learning and transfer learning is proposed. This algorithm is based on the Densely Connected Convolutional Networks (DenseNet) structure of deep neural networks, and constructs a network model by introducing attention mechanisms, and trains the enhanced dataset using multi-level transfer learning. Experimental results demonstrate that the algorithm achieves an efficiency of over 84.0% in the test set, with a significantly improved classification accuracy compared to previous models, making it applicable to medical breast cancer detection tasks.

With the fast growth of web, the web is full of diverse data. Linked Data is data on the web that gives URIs to entities and links different data from different domains together, which results in the part of the web infrastructure. Since the dramatic development of Linked Data movement, a great number of accessibility datasets have been published as open data on the web by governments or organizations so that they can be accessed and reused by anyone without any restriction. However, the majority of the accessibility datasets available on the web are in diverse formats such as HTML table, Excel and CSV, rather than Linked Data format – RDF. This paper mainly investigates the method of creating accessibility Linked Data based on the publicly existing available accessibility datasets. First of all, requested accessibility data was extracted from the accessibility dataset. Moreover, Jena RDF API is applied to create RDF documents based on the extracted data from the accessibility dataset. Compared to other approaches such as transforming XML to RDF via schema, current method of this project does not have mapping work and it shows significant convenience. Finally, a triple store was set up to sever the RDF documents. In addition, the server has query interface support for test; as a result, users could query the data they need. In the future development, relevant dataset links would be added to current RDF documents. At last the accessibility RDF dump would be published on the web as a part of the web infrastructure. Index Terms—Linked Data, Open Data, Semantic Web, Accessibility Dataset.

Small target detection is an important research direction in computer vision, widely applied in scenarios such as drone monitoring, remote sensing image analysis, and autonomous driving. However, as small targets occupy fewer pixels, contain limited feature information, and often appear in complex backgrounds, existing detection algorithms face shortcomings in accuracy and robustness. To address this, this paper proposes a novel small target detection algorithm that integrates hierarchical feature fusion with a spatial dynamic collaboration mechanism. The hierarchical feature fusion module (HFA) effectively combines shallow detail features with deep semantic features, greatly enhancing the feature representation capability for small targets. Meanwhile, the dynamic collaboration mechanism (DCCA) dynamically adjusts feature fusion weights and detection strategies based on target scale and density distribution, thereby further improving detection accuracy and robustness. Extensive experiments are conducted on datasets such as VisDrone, TinyPerson, and NWPU VHR-10. Results demonstrate that, compared to state-of-the-art models like YOLOv8 and YOLOv10, the proposed algorithm achieves significant improvements in precision, recall, and mAP, with mAP increasing by 2.1% to 3.2% and mAP-95 by 1.2% to 1.8%. Ablation studies further validate the complementarity of HFA and DCCA in optimizing model performance, confirming the algorithm’s superiority and robustness in complex scenarios. This research provides a novel technical route for small target detection and offers valuable references for practical applications.

This paper delves into the significance and interaction between deep learning (DL) and neural architecture search (NAS) within the realm of artificial intelligence. As DL has become integral in addressing complex problems through its robust learning capabilities, NAS offers the potential to autonomously discover efficient neural network structures. The fusion of these technologies not only enhances model performance but also accelerates the proliferation of AI applications, particularly in resource-constrained settings such as embedded devices and edge computing. This study provides an in-depth comparative analysis of multiple neural network models applied to the CIFAR-10 dataset, with a particular focus on the performance of the Darts-SEBnet model. By incorporating a self-attention mechanism, the Darts-SEBnet model demonstrates a significant improvement in accuracy over the baseline VGG16 model. Furthermore, the paper reviews the evolution of NAS, emphasizing the success of gradient-based search methods like DARTS and its variants in improving search efficiency and model performance. The findings suggest that the integration of DL and NAS could drive further innovations in AI, offering solutions to existing bottlenecks and expanding AI's applicability across diverse fields.