OBJECTIVES: Anaphora recognition is a process to identify exactly which noun has been used previously and relates to a pronoun that is included in a specific sentence later. Therefore, anaphora recognition is an essential element of a dialogue agent system. In the current study, all the merits of rule-based, machine learning-based, semantic-based anaphora recognition systems were combined to design and realize a new hybrid-type anaphora recognition system with an optimum capacity. METHODS: Anaphora recognition rules were encoded on the basis of the internal traits of referred expressions and adjacent contexts to realize a rule-based system and to serve as a baseline. A semantic database, related to predicate instances of sentences including referred expressions, was constructed to identify semantic co-relationships between the referent candidates (to which semantic tags were attached) and the semantic information of predicates. This approach would upgrade the anaphora recognition system by reducing the number of referent candidates. Additionally, to realize a machine learning-based system, an anaphora recognition model was developed on the basis of training data, which indicated referred expressions and referents. The three methods were further combined to develop a new single hybrid-based anaphora recognition system. RESULTS: The precision rate of the rule-based systems was 54.9%. However, the precision rate of the hybrid-based system was 63.7%, proving it to be the most efficient method. CONCLUSIONS: The hybrid-based method, developed by the combination of rule-based and machine learning-based methods, represents a new system with enhanced functional capabilities as compared to other pre-existing individual methods.
Delivery of Health Care* ; Natural Language Processing ; Semantics

PURPOSE: The aim of this study was to investigate the clinical characteristics and causative organisms of meningitis in the Daegu region and seek a useful tool for the early prediction of bacterial meningitis in children. METHODS: We retrospectively reviewed the medical records of 115 pediatric patients diagnosed with bacterial or aseptic meningitis at Yeungnam university hospital in Daegu from March 2012 to July 2013. We evaluated their clinical symptoms, laboratory findings, clinical courses, bacterial meningitis scores and complications. RESULTS: The subjects included 106 with aseptic meningitis and 9 with bacterial meningitis. At the time of visit, fever was the most frequent symptom, followed by headache, vomiting and neck stiffness. In cerebrospinal fluid (CSF) analysis, white blood cell (WBC) count were higher in the bacterial meningitis group (1423.8+/-1980.4 vs. 120.0+/-161.6 mg/dL). Mean CSF protein was 219.4+/-183.6 mg/dL in bacterial meningitis and 42.4+/-27.0 mg/dL in aseptic meningitis (P <0.001). Bacterial meningitis score (BMS) were higher in the group with bacterial meningitis. Abnormal radiological findings were found in 44% of the group with bacterial meningitis. CONCLUSION: Although the clinical features between the groups were similar, the CSF analysis revealed significant differences statistically. Furthermore, BMS could be helpful to predict bacterial meningitis in children. During the outbreak of aseptic meningitis, it might reduce unnecessary hospital admissions and antibiotic treatments.
Cerebrospinal Fluid ; Child ; Daegu ; Fever ; Headache ; Humans ; Leukocytes ; Medical Records ; Meningitis ; Meningitis, Aseptic* ; Meningitis, Bacterial ; Neck ; Retrospective Studies ; Vomiting