데이터기반 입법수요 예측 모형 개발 및 고도화

제1장 서론 제1절 연구의 배경 및 목적 ····································································································3 1. 연구 배경 ······························································································································3 2. 연구 내용 및 활용 방안 ·······································································································3 제2장 언론데이터에서 법제･입법 이슈와 관련성 판단 및 분류를 위한 연구 제1절 연구개요 ·······················································································································7 1. 들어가며 ································································································································7 2. 선행 연구 ······························································································································8 3. 연구 내용과 연구 목표 ·········································································································9 4. 텍스트 문서 분류 방법론 ······································································································9 가. SVM ·····································································································································10 나. Linear Discriminant Analysis ···························································································11 다. 로지스틱 회귀분석 ·················································································································11 라. 나이브 베이즈 ·······················································································································12 마. 랜덤 포레스트(Random Forest) ··························································································12 바. AdaBoost ·····························································································································13 사. XGBoost ······························································································································14 제2절 언론데이터 전처리 및 불용어 처리 기준 ···································································15 1. 목표 모집단과 조사 모집단 ································································································15 가. 목표 모집단과 조사 모집단 ···································································································15 나. 학습 및 테스트 데이터 ··········································································································17 2. 데이터 전처리 ·····················································································································18 가. 빅카인즈 언론데이터 ·············································································································18 제3절 분류 방법론의 적용과 결과 ························································································23 경제⋅인문사회연구회1. 모델별 성능 ························································································································23 제3장 국회 입법데이터에서 내용기반의 분류기준 마련 연구 제1절 토픽모델링의 개념 및 분석 방법론 ············································································31 1. 토픽모델링의 개념 ··············································································································31 2. 분석 목적 및 방법론 ··········································································································31 제2절 LDA를 이용한 국회 입법데이터의 내용기반 분류기준 ···············································34 1. 분석 설정 ····························································································································34 2. 분석 결과와 해석: 지역 등 중점이슈 ··················································································34 제3절 HDP를 이용한 국회 입법데이터의 내용기반 분류기준 ··············································58 1. 분석 설정 ····························································································································58 2. 분석 결과와 해석: 지역 등 중점이슈, LDA와의 비교를 중심으로 ·····································58 제4장 (pilot study)…

출처: 국립중앙도서관 서지정보