NAIST QA System for QAC2
نویسندگان
چکیده
The system we presented for the subtask1 and subtask2 in QAC2 is based on our previous one [12], which utilized a greedy answer seeking model using paraphrasing. We incorporate a re-ranking model for matching questions and passages into the previous system. In the model, we integrated a proximitybased scoring function with the structural-based scoring function. Unfortunately, the result of evaluation shows that our proposed model did not work well. Based on error analysis, we conclude that structural matching-based approaches to answer seeking require technologies for large-scale acquisition of paraphrase patterns. We are now investigating a variation of paraphrasing which is expected to be helpful for question answering.
منابع مشابه
Question Answering System QUARK
Recently, we can acquire immence amount of information thanks to the spread of a computer and internet. Therefore, technology for finding the information that a user desires becomes more and more important. A question answering (QA) system answers a question written by natural language in contrast to conventional information retrieval systems where a user expresses his information need by keywo...
متن کاملNTT's Question Answering System for NTCIR QAC2
In order to retrieve best documents for finding answers, we developed a robust proximity search engine. It efficiently finds relevant passages. In addition, our search engine has two disjunction operators: or and or2. The former works just like addition, whereas the latter works just like logical disjunction. The operator or2 is used to introduce synonyms and antonyms of a query term. The searc...
متن کاملToshiba ASKMi at NTCIR-4 QAC2
Toshiba participated in NTCIR-4 QAC2 Subtask 1: this is our first QAC participation. Using our newly developed Japanese QA system called ASKMi, we submitted two runs, one using Okapi/BM25 for document retrieval (TSB-A) and the other using Boolean AND constraints before applying Okapi/BM25 (TSB-B). We achieved the 5th best performance among the 17 participants (8th and 9th among the 25 submitted...
متن کاملNTCIR-4 QAC Experiments at Matsushita
This paper investigates our experimental results for NTCIR-4 QAC2, the second attempt to evaluate the technology of Japanese question answering (QA). Our basic approach is a combination of information retrieval and named entity (NE) extraction based on pattern matching. The results show that the accuracy of NE extraction crucially affects the overall performance of our system. Additional experi...
متن کاملQuestion Answering Challenge for Five Ranked Answers and List Answers - Overview of NTCIR4 QAC2 Subtask 1 and 2
In this paper we describe an evaluation of question answering task, Question Answering Challenge 2 (QAC2). This evaluation project was first carried out at the NTCIR Workshop 3 in October 2002. One objective of the QAC was to develop practical QA systems in a general domain by focusing on research relating to user interaction and information extraction. Our second objective was to develop an ev...
متن کامل