Risk Commodity Dataset (RCDD) from Alibaba's e-commerce platform

This is a risk commodity detection dataset (RCDD) that is based on a real risk detection scenario from Alibaba's e-commerce platform. All the <em>.csv </em>files are the initial data which consists of edges, node features, supervised information as well as IDs of candidate items. More details can see in <em>README.md</em>, we list the type of each column in each file as follows: 1. RCDD_edges.csv (edge file): <br> source_node_id int, target_node_id int, source_node_type string, target_node_type string, edge_type string<br> 2. RCDD_nodes.csv (node file):<br> node_id int, node_type string, node_atts string (notice that node_atts are 256-dimensional feature vector strings with delimiter ":")<br> 3. RCDD_train_labels.csv (training labels):<br> item_id int, label int<br> 4. RCDD_test_ids.csv (testing ids):<br> item_id int<br> 5. RCDD_test_labels.csv (testing labels):<br> item_id int, label int Besides that,<em> graph.bin</em> is the format in DGL which is constructed by all <em>*.csv</em> files, and a general method to load this graph as follows: <pre><code class="language-python">from dgl import load_graphs #should install dgl ds,_ = load_graphs("./graph.bin") g = ds[0] print(g)</code></pre> And then you can easily get a large-scale heterogeneous graph with 157,814,864 edges and 13,806,619 nodes, our graph task is node classification: detect risk product.风险商品检测数据集（Risk Commodity Detection Dataset, RCDD）基于阿里巴巴电商平台的真实风险检测场景构建。所有.csv格式文件为初始数据，涵盖边信息、节点特征、监督标签以及候选商品ID。更多细节可参见README.md，下文列出各文件中每一列的数据类型： 1. RCDD_edges.csv（边文件）：source_node_id（整型）、target_node_id（整型）、source_node_type（字符串型）、target_node_type（字符串型）、edge_type（字符串型） 2. RCDD_nodes.csv（节点文件）：node_id（整型）、node_type（字符串型）、node_atts（字符串型，注：node_atts为以冒号分隔的256维特征向量字符串） 3. RCDD_train_labels.csv（训练标签集）：item_id（整型）、label（整型） 4. RCDD_test_ids.csv（测试ID集）：item_id（整型） 5. RCDD_test_labels.csv（测试标签集）：item_id（整型）、label（整型） 此外，graph.bin为深度图库（Deep Graph Library, DGL）格式文件，由所有*.csv文件构建而成，加载该图的通用方法如下： python from dgl import load_graphs # 需先安装DGL库 ds,_ = load_graphs("./graph.bin") g = ds[0] print(g) 该异构图规模庞大，包含157,814,864条边与13,806,619个节点，本数据集的图任务为节点分类任务：风险商品检测。