Character Mining Project Sample Clauses

Character Mining Project. The Character Mining 1 dataset provides transcripts of the TV show Friends as well as annotation for several tasks. Future research could combine this project 1xxxxx://xxxxxx.xxx/emorynlp/character-mining with FriendsQA to generate more meaningful tasks and tools. The first two seasons are annotated [3] for character identification task, that is an entity linking task identifying personal mentions with character names. This annotation is extended to the next two seasons and ambiguous mentions are further annotated[2]. Build- ing upon that, plural mentions of those four seasons [31] are also annotated for character identification tasks. Moreover, the first four seasons are annotated [30] for fine-grained emotion detection tasks. Further more, selected dialogues from all ten seasons are processed [14] for a cloze-style reading comprehension task. Besides, a personality detection task was annotated by Jiang et al. [9]. Finally a span based question answering task was annotated by Yang et al. [29] Chapter 3 Approach Softmax oµ ij ec es ew … ew … eµ … es ew … ew