计算机应用研究
計算機應用研究
계산궤응용연구
APPLICATION RESEARCH OF COMPUTERS
2015年
2期
547-551
,共5页
动态话题模型%话题追踪%误报检测%信念网络
動態話題模型%話題追蹤%誤報檢測%信唸網絡
동태화제모형%화제추종%오보검측%신념망락
dynamic topic model%topic tracking%error detection%belief network
针对动态话题追踪模型高误报率的现象,提出了动态追踪中的误报检测来判断追踪到的相关报道是否误报,进而降低动态模型的误报率。考虑到新报道是否和话题相关,除了依据两者的相似度外,还涉及时间距离、差值关系、分布关系、追踪到的报道和话题核心报道的相似度四方面内容,给出了误报检测因子计算式。实验采用TDT4测试集合和DET曲线进行评测,通过反复实验获得了误报检测因子δ的阈值,与基于信念网络的动态话题追踪模型相比,使用误报检测后模型的最优(Cdet )norm降低了5.032%。
針對動態話題追蹤模型高誤報率的現象,提齣瞭動態追蹤中的誤報檢測來判斷追蹤到的相關報道是否誤報,進而降低動態模型的誤報率。攷慮到新報道是否和話題相關,除瞭依據兩者的相似度外,還涉及時間距離、差值關繫、分佈關繫、追蹤到的報道和話題覈心報道的相似度四方麵內容,給齣瞭誤報檢測因子計算式。實驗採用TDT4測試集閤和DET麯線進行評測,通過反複實驗穫得瞭誤報檢測因子δ的閾值,與基于信唸網絡的動態話題追蹤模型相比,使用誤報檢測後模型的最優(Cdet )norm降低瞭5.032%。
침대동태화제추종모형고오보솔적현상,제출료동태추종중적오보검측래판단추종도적상관보도시부오보,진이강저동태모형적오보솔。고필도신보도시부화화제상관,제료의거량자적상사도외,환섭급시간거리、차치관계、분포관계、추종도적보도화화제핵심보도적상사도사방면내용,급출료오보검측인자계산식。실험채용TDT4측시집합화DET곡선진행평측,통과반복실험획득료오보검측인자δ적역치,여기우신념망락적동태화제추종모형상비,사용오보검측후모형적최우(Cdet )norm강저료5.032%。
For the reason of high false alarm probability in dynamic topic tracking,this paper proposed error detection to judge whether the tracked related story was belong to false alarm and then decreased false alarm probability in dynamic topic trac-king.Considering a new story was whether related to a topic,not only depending on the similarity between the two,but also de-pending on difference relationship,distribution relationship,and the similarity between new story and seminal story of existed topic,gave the computation formula of error detection.TDT4 corpora and DET curves were used to run experiments.This paper firstly obtained the threshold of error detection factorδ,the tracking performance of dynamic topic model based on belief net-work decreases (Cdet )norm by 5 .032% when uses error detection.