XJNLP at SemEval-2017 Task 12: Clinical temporal information ex-traction with a Hybrid Model

000 001 002 003 00 004 005 006 007 008 09 010 011 012 013 014 015 016 017 018 019 020 021 022 023 024 025 026 027 028 029 030 031 032 033 034 035 036 037 038 039 040 041 042 043 044 045 046 047 048 049 050 051 052 053 054 055 056 057 058 059 060 061 062 063 064 065 066 067 068 069 070 071 072 073 074 075 076 077 078 079 080 081 082 083 084 085 086 087 088 089 090 091 092 093 094 095 096 097 098 XJNLP at SemEval-2017 Task 12: Clinical temporal information ex- traction with a Hybrid Model Yu Long Zhijing Li Xi’an Jiaotong University Xi’an Jiaotong University yulongxbdx@163.com tokyojackson@stu.xjtu.edu.cn Xuan Wang Chen Li Xi’an Jiaotong University Xi’an Jiaotong University m18829236255@163.com cli@xjtu.edu.cn Abstract Temporality is crucial in understanding the course of clinical events from a patient’s electronic health records and temporal pro- cessing is becoming more and more im- portant for improving access to content. SemEval 2017 Task 12 (Clinical TempEval) addressed this challenge using the THYME corpus, a corpus of clinical narratives anno- tated with a schema based on TimeML2 guidelines. We developed and evaluated ap- proaches for: extraction of temporal expres- sions (TIMEX3) and EVENTs; EVENT at- tributes; document-time relations. Our ap- proach is a hybrid model which is based on rule based methods, semi-supervised learn- ing, and semantic features with addition of manually crafted rules. 1 Introduction Extraction and interpretation of temporal infor- mation from clinical text is essential for clinical practitioners and researchers. Extracting temporal information from unstructured clinical narratives is an important step towards the accurate construc- tion of a patient timeline over the course of clinical care. SemEval-2017 Task 12 (Clinical TempEval) is a direct successor to 2016 Clinical TempEval. Clinical TempEval is designed to ad-dress the challenge of understanding clinical timeline in medical narratives and it is based on the THYME corpus which includes temporal an-notations. Researchers have explored ways to extract tem- poral information from clinical text. Lee et al. (2016) developed an approach based on linear and structural (HMM) support vector machines using lexical, morphological, syntactic, discourse, and word representation features. P R, Sarath et al. (2016) used a hybrid approach(rule-based and ma- chine learning) for temporal information extraction from clinical notes. Velupillai et al. (2015) devel- oped a pipeline based on ClearTK and SVM with lexical features to extract TIMEX3 and EVENT mentions. Most of the participants of these chal- lenges used CRF and SVM for event and time ex- pression extraction with features including the in- formation gathered from different resources like UMLS (Unified Medical Language System), out- put of TARSQI toolkit, Brown Clustering, Wikipe- dia and Metamap (Aronson and Lang, 2010). Those machine-learning methods are complex and they cost much time to run. However, they can be not only flexible but also convenient when com- pared to the handcrafting label. Others also used some rule based methods, which are fast but not …

Content truncated to 500 words