M-Evolve: Structural-Mapping-Based Data Augmentation for Graph Classification
نویسندگان
چکیده
Graph classification, which aims to identify the category labels of graphs, plays a significant role in drug toxicity detection, protein analysis etc. However, limitation scale benchmark datasets makes it easy for graph classification models fall into over-fitting and undergeneralization. To improve this, we introduce data augmentation on graphs (i.e. augmentation) present four methods:random mapping, vertex-similarity motif-random mapping motif-similarity generate more weakly labeled small-scale via heuristic transformation structures. Furthermore, propose generic model evolution framework, named M-Evolve, combines augmentation, filtration retraining optimize pre-trained classifiers. Experiments six demonstrate that proposed framework helps existing alleviate undergeneralization training datasets, successfully yields an average improvement 3 - 13% accuracy tasks.
منابع مشابه
Data Augmentation for Plant Classification
Data augmentation plays a crucial role in increasing the number of training images, which often aids to improve classification performances of deep learning techniques for computer vision problems. In this paper, we employ the deep learning framework and determine the effects of several data-augmentation (DA) techniques for plant classification problems. For this, we use two convolutional neura...
متن کاملAnatomical Data Augmentation For CNN based Pixel-wise Classification
In this work we propose a method for anatomical data augmentation that is based on using slices of computed tomography (CT) examinations that are adjacent to labeled slices as another resource of labeled data for training the network. The extended labeled data is used to train a U-net network for a pixel-wise classification into different hepatic lesions and normal liver tissues. Our dataset co...
متن کاملM - Infosift : a Graph - Based Approach for Multiclass Document Classification
M-INFOSIFT: A GRAPH-BASED APPROACH FOR MULTICLASS DOCUMENT CLASSIFICATION
متن کاملReal Data Augmentation for Medical Image Classification
Many medical image classification tasks share a common unbalanced data problem. That is images of the target classes, e.g., certain types of diseases, only appear in a very small portion of the entire dataset. Nowadays, large co llections of medical images are readily available. However, it is costly and may not even be feasible for medical experts to manually comb through a huge unlabeled data...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: IEEE Transactions on Network Science and Engineering
سال: 2021
ISSN: ['2334-329X', '2327-4697']
DOI: https://doi.org/10.1109/tnse.2020.3032950