Add Multilingual Poetry Subject Clustering Benchmark - 题材 Clustering Benchmark

README.md

@@ -1,3 +1,57 @@
 ---
-license: cc-by-4.0
+configs:
+- config_name: default
+  data_files:
+  - path: test-00000-of-00001.parquet
+    split: test
+dataset_info:
+  config_name: default
+  dataset_size: 4463765
+  download_size: 4463765
+  features:
+  - dtype: string
+    name: id
+  - dtype: string
+    name: author
+  - dtype: string
+    name: title
+  - dtype: string
+    name: text
+  - dtype: string
+    name: subject
+  - dtype: string
+    name: subject_code
+  - dtype: float64
+    name: score
+  - dtype: string
+    name: language
+  splits:
+  - name: test
+    num_bytes: 4463765
+    num_examples: 8904
 ---
+
+# Multilingual Poetry Subject Clustering Benchmark
+
+## 数据集描述
+
+这是一个多语种诗歌题材聚类基准，用于评估跨语言表示模型在无监督诗歌聚类任务上的性能。每条记录包含一首诗歌及其所属的题材聚类标签，以及它与聚类中心向量的相似度分数（余弦相似度）。
+
+## 数据集统计
+
+| 指标 | 值 |
+|------|----------|
+| **诗歌总数** | 8,904 首诗歌 |
+| **文件大小** | 4.3 MB |
+| **文件名称** | test-00000-of-00001.parquet |
+
+## 数据字段说明
+
+- `id`: 诗歌唯一标识符
+- `author`: 诗歌作者
+- `title`: 诗歌标题  
+- `text`: 诗歌正文内容
+- `language`: 诗歌语言代码
+- `subject`: 诗歌题材分析标签 (原始数据)
+- `subject_code`: 聚类中心分配的题材标签 (用于聚类评估的基准)
+- `score`: 该诗歌与其聚类中心向量的余弦相似度（用于衡量聚类质量）

dataset_infos.json

@@ -0,0 +1,53 @@
+{
+  "default": {
+    "description": "Multilingual Poetry Subject Clustering Benchmark - default config",
+    "citation": "",
+    "homepage": "",
+    "license": "apache-2.0",
+    "features": {
+      "id": {
+        "dtype": "string",
+        "_type": "Value"
+      },
+      "author": {
+        "dtype": "string",
+        "_type": "Value"
+      },
+      "title": {
+        "dtype": "string",
+        "_type": "Value"
+      },
+      "text": {
+        "dtype": "string",
+        "_type": "Value"
+      },
+      "subject": {
+        "dtype": "string",
+        "_type": "Value"
+      },
+      "subject_code": {
+        "dtype": "string",
+        "_type": "Value"
+      },
+      "score": {
+        "dtype": "float64",
+        "_type": "Value"
+      },
+      "language": {
+        "dtype": "string",
+        "_type": "Value"
+      }
+    },
+    "splits": {
+      "test": {
+        "name": "test",
+        "num_bytes": 4463765,
+        "num_examples": 8904,
+        "dataset_name": "MultilingualPoetrySubjectClustering"
+      }
+    },
+    "download_size": 4463765,
+    "dataset_size": 4463765,
+    "size_in_bytes": 4463765
+  }
+}

test-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bfef5ad2f96cdad73502230ebe3cd38c4cc583d3dd357c3d6b62dfb1834af253
+size 4463765