A Chinese Long-text Corpus for Entity Linking
There are three directories： Texts(Raw texts of news and commentaries), Anno_zhwiki(Annotation to Chinese Wikipedia), Anno_DB(Annotation to CN-DBpedia).
As for the format of annotation files, it follows the structure of: "Mention/StartPosition/EndPosition/Wikipeida/CN-DBpedia Identifier" The items are split by tab.
Also, there might be potential mistakes. We will keep them updated.
Moreover, the dataset is also divided into three categories according to the difficulty: Easy (32): 1 2 6 8 9 12 17 20 22 24 27 28 32 33 35 41 42 43 47 63 71 74 75 77 80 81 86 87 94 98 99 100
Medium (49): 3 4 5 7 11 14 16 18 19 23 25 26 31 34 36 37 45 46 48 49 50 52 53 55 58 59 60 61 66 67 68 69 72 73 76 78 79 82 83 84 85 88 89 90 91 92 93 95 96
Hard (19): 10 13 15 21 29 30 38 39 40 44 51 54 56 57 62 64 65 70 97