add DBLP-4K dataset

Author: yifanfeng97

dhg/data/__init__.py

@@ -8,7 +8,7 @@
 from .white_house import HouseCommittees
 from .news import News20
 from .coauthorship import CoauthorshipCora, CoauthorshipDBLP
-from .dblp import DBLP8k
+from .dblp import DBLP4k, DBLP8k
 from .cocitation import CocitationCora, CocitationCiteseer, CocitationPubmed
 from .blogcatalog import BlogCatalog
 from .flickr import Flickr
@@ -39,6 +39,7 @@
     "PubmedBiGraph",
     "CoauthorshipCora",
     "CoauthorshipDBLP",
+    "DBLP4k",
     "DBLP8k",
     "CocitationCora",
     "CocitationCiteseer",

dhg/data/dblp.py

@@ -1,6 +1,7 @@
 from typing import Optional
+from functools import partial
 
-from dhg.datapipe import load_from_pickle
+from dhg.datapipe import load_from_pickle, norm_ft, to_bool_tensor, to_tensor, to_long_tensor
 
 from .base import BaseData
 
@@ -29,3 +30,74 @@ def __init__(self, data_root: Optional[str] = None) -> None:
             },
         }
 
+
+class DBLP4k(BaseData):
+    r"""The DBLP-4k dataset is a citation network hypergraph dataset for node classification task.
+    The dataset is an academic network from four research areas. There are 14,475 authors, 
+    14,376 papers, and 20 conferences, among which 4,057 authors, 20 conferences and 
+    100 papers are labeled with one of the four research areas (database, data mining, machine learning, and information retrieval). 
+    The vertice denotes author, and three types of correlation (co-paper, co-term, co-conference) can be used for building hyperedges.
+    More details see the `PathSim: Meta Path-Based Top-K Similarity Search in Heterogeneous Information Networks <http://www.vldb.org/pvldb/vol4/p992-sun.pdf>`_ paper.
+    
+    The content of the DBLP-4k dataset includes the following:
+
+    - ``num_classes``: The number of classes: :math:`4`.
+    - ``num_vertices``: The number of vertices: :math:`4,057`.
+    - ``num_paper_edges``: The number of hyperedges constructed by the co-paper correlation: :math:`14,328`.
+    - ``num_term_edges``: The number of hyperedges constructed by the co-term correlation: :math:`7,723`.
+    - ``num_conf_edges``: The number of hyperedges constructed by the co-conference correlation: :math:`20`.
+    - ``dim_features``: The dimension of author features: :math:`334`.
+    - ``features``: The author feature matrix. ``torch.Tensor`` with size :math:`(4,057 \times 334)`.
+    - ``labels``: The label list. ``torch.LongTensor`` with size :math:`(4,057, )`.
+    - ``edge_by_paper``: The hyperedge list constructed by the co-paper correlation. ``List`` with length :math:`(14,328 \times 2)`.
+    - ``edge_by_term``: The hyperedge list constructed by the co-term correlation. ``List`` with length :math:`(7,723 \times 2)`.
+    - ``edge_by_conf``: The hyperedge list constructed by the co-conference correlation. ``List`` with length :math:`(20 \times 2)`.
+    - ``paper_author_dict``: The dictionary of ``{paper_id: [author_id, ...]}``. ``Dict`` with length :math:`(14,328)`.
+    - ``term_paper_dict``: The dictionary of ``{term_id: [paper_id, ...]}``. ``Dict`` with length :math:`(7,723)`.
+    - ``conf_paper_dict``: The dictionary of ``{conf_id: [paper_id, ...]}``. ``Dict`` with length :math:`(20)`.
+    """
+
+    def __init__(self, data_root: Optional[str] = None):
+        super().__init__("dblp_4k", data_root)
+        self._content = {
+            'num_classes': 4,
+            'num_vertices': 4057,
+            'num_paper_edges': 14328,
+            'num_term_edges': 7723,
+            'num_conf_edges': 20,
+            'dim_features': 334,
+            "features": {
+                "upon": [{"filename": "features.pkl", "md5": "7f8e6c3219026c284342d45c01e16406"}],
+                "loader": load_from_pickle,
+                "preprocess": [to_tensor, partial(norm_ft, ord=1)],
+            },
+            'labels': {
+                'upon': [{'filename': 'labels.pkl', 'md5': '6ffe5ab8c5670d8b5df595b5c4c63184'}],
+                'loader': load_from_pickle,
+                'preprocess': [to_long_tensor]
+            },
+            'edge_by_paper': {
+                'upon': [{'filename': 'edge_by_paper.pkl', 'md5': 'e473eddeb4692f732bc1e47ae94d62c2'}],
+                'loader': load_from_pickle,
+            },
+            'edge_by_term': {
+                'upon': [{'filename': 'edge_by_term.pkl', 'md5': '1ca7cfbf46a7f5fc743818c65392a0ed'}],
+                'loader': load_from_pickle,
+            },
+            'edge_by_conf': {
+                'upon': [{'filename': 'edge_by_conf.pkl', 'md5': '890d683b7d8f943ac6d7e87043e0355e'}],
+                'loader': load_from_pickle,
+            },
+            'paper_author_dict': {
+                'upon': [{'filename': 'paper_author_dict.pkl', 'md5': 'eb2922e010a78961b5b66e77f9bdf950'}],
+                'loader': load_from_pickle,
+            },
+            'term_paper_dict': {
+                'upon': [{'filename': 'term_paper_dict.pkl', 'md5': '1d71f988b52b0e1da9d12f1d3fe24350'}],
+                'loader': load_from_pickle,
+            },
+            'conf_paper_dict': {
+                'upon': [{'filename': 'conf_paper_dict.pkl', 'md5': 'cbf87d64dce4ef40d2ab8406e1ee10e1'}],
+                'loader': load_from_pickle,
+            },
+        }

docs/source/api/data.rst

@@ -61,6 +61,7 @@ Hypergraph Datasets
     dhg.data.WalmartTrips
     dhg.data.HouseCommittees
     dhg.data.News20
+    dhg.data.DBLP4k
     dhg.data.DBLP8k