본문 바로가기 주메뉴 바로가기
국회도서관 홈으로 정보검색 소장정보 검색

결과 내 검색

동의어 포함

목차보기

Title page 1

Contents 1

Abstract 2

I. Introduction 4

II. Literature Review 5

III. Methodology 6

IV. Results Validation 8

1. Internal Validation 8

2. External Validation 11

V. Key Findings 12

1. Emerging Areas Not Tracked in CRS 12

2. Unpacking Multi-Sectoral Aid 13

3. Unpacking Aid Unallocated by Sector 14

4. Estimating Flows to Cross-cutting and Thematic Priorities 15

VI. Conclusion 17

References 18

Appendix: Supplementary Methodological Notes 21

I. Pilot Phase and ML-Framework Selection 21

II. Machine-Generated Clusters 23

III. Mapping of Clusters to Thematic Areas 25

Tables 12

Table 1. Comparison of ML Generated Sectors with IDA Commitments Data 12

Table 2. Top 10 ML-Cluster Destination for CRS "Sectors not Specified" records 15

Table 3. Mapping of ML-Generated Clusters to IDA20 Special Theme "Climate Change" 16

Figures 7

Figure 1. Snapshot of ML Methodology to Process OECD CRS Data 7

Figure 2. Frequency distribution of cosine similarity scores between two rounds of AI/ML clustering 9

Figure 3. Comparison of estimation of CRS Sector Groups using ML-Generated Clusters to Original CRS Sector Groups 2021 data, $US Millions in 2021 prices 10

Figure 4. Methodology for creating sub-sectors 13

Figure 5. ML-generated Multi-Sector/Cross Cutting Sub-Sectors In $US millions in 2021 prices 14

Figure 6. Estimation of 2021 Commitments to WB GCPs, IDA20 Special Themes, and IDA20 Cross-Cutting Issues In $US billions in 2021 prices 17

Appendix Tables 21

Table A.1. Considered Machine Learning Frameworks 21

Table A.2. Components of Chosen Machine Learning Framework - Approach 2 (Fast Clustering) 21

Table A.3. Key Parameters for Fast Clustering Algorithm 23

Table A.4. Newly Created Sub-Sectors 24