earticle

논문검색

Prediction of Research Topics on Science & Technology (S&T) using Ensemble Forecasting

초록

영어

Proper resource allocation on research requires accurate forecasting for the future research activities. Forecasting task can be done using judgmental or numerical analysis. Bibliometric analysis is a quantitative method to determine the trend of research area by counting the frequency of certain keywords using journal publication or patents. This paper reports the implementation of our new forecast combination method which selects the best methods used by similar validation dataset on Indonesian journal database, namely the Garuda dataset, especially on the subject of Science and Technology. The experimental result indicates that the proposed method may perform better compared to the fix combination of predictors. In addition, based on the prediction result, the emerging research topics for the next few years can be objectively identified.

목차

Abstract
 1. Introduction
 2. Theoretical Background
  2.1. Forecast Combinations
  2.2. Model Selection
  2.3. Time Series Similarity
  2.3. Growth Rate
 3. Experimental Setup
  3.1. Methodology
  3.2. Datasets
  3.3. Performance Evaluation
  3.4. Hardware and Tools
 4. Result and Discussion
  4.1. Comparison among Individual Predictor
  4.2. Combination of Models using Similarity Measure
  4.3. Emerging Topics
 4. Conclusion
 References

저자정보

  • Indra Budi Faculty of Computer Science University of Indonesia
  • Rizal Fathoni Aji Faculty of Computer Science University of Indonesia
  • Agus Widodo Faculty of Computer Science University of Indonesia

참고문헌

자료제공 : 네이버학술정보

    함께 이용한 논문

      ※ 원문제공기관과의 협약기간이 종료되어 열람이 제한될 수 있습니다.

      0개의 논문이 장바구니에 담겼습니다.