DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Lee, Chang-Ock | - |
dc.contributor.advisor | 이창옥 | - |
dc.contributor.author | Jung, Young-Han | - |
dc.date.accessioned | 2019-09-03T02:44:52Z | - |
dc.date.available | 2019-09-03T02:44:52Z | - |
dc.date.issued | 2018 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=828529&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/266410 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : 수리과학과, 2018.8,[iii, 13 p. :] | - |
dc.description.abstract | With the advancement of technology and big data, machine learning has become the trend in a lot of researches. This is due to the high eciency of machine learning. This thesis presents an implementation of recurrent neural network (RNN) model, a type of model of machine learning, for solving math word problem. We use seq2seq model to increase the accuracy of the performance. However, despite the decrease in the cost function, the accuracy of the machine is very low. Thus, we consider the limitations of our model that could be improved. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | Deep learning▼agated recurrent unit▼along short-term memory▼amath word problem▼arecurrent neural network | - |
dc.subject | 심층 학습▼aGated recurrent unit▼along short-term memory▼a문장형 수학 문제▼a순환신경망 | - |
dc.title | (A) study on recurrent neural network model by solving math word problems | - |
dc.title.alternative | 문장형 수학문제 풀이 모델을 이용한 순환신경망 탐구 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :수리과학과, | - |
dc.contributor.alternativeauthor | 정영한 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.