Browsing by Author "Hsu, Ping-Yu"

Now showing 1 - 2 of 2

A Boosting Regression-Based Method to Evaluate the Vital Essence in Semiconductor Industry Performance
(2020-01-01)
Hsu, Ping-Yu
;
Yeh, I-Wen
;
Tseng, Ching-Hsun
;
Lee, Shin-Jye
;
科技管理研究所
In accordance with the statistical analysis, the industrial performance is usually related to research and development (R&D) intensity, and this factor indeed plausibly brings the biggest profit with patents and supporting products to the development of semiconductor industry. How to evaluate the completive performance of modern industries is an increasing issue, especially for the semiconductor industries in these decades. However, almost every traditional statistical model is deterred by the hypothesis of population and independent correlation among each feature, and this makes the result of typical regression model potentially lose reliability. To avoid this weakness, this article therefore applies a gradient boosting based method - XGBoost to evaluate the feature importance of semiconductor industries. In the simulation experiments, different findings revel certain information, apart from R&D intensity, actually sway the gross net value in the annual financial announcement of semiconductor industries. Moreover, this article proposes another concept to evaluate the essential factor contributing the development of semiconductor industries. Instead of only focusing on the effect of R&D intensity, this article also predicts the future growth rate (GR) of net value by applying the greedy search of XGBoost Regression.
An approach to discover and recommend cross-domain bridge-keywords in document banks
(2010)
Su, Yu-Min
;
Hsu, Ping-Yu
;
Pai, Ning-Yao
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
Purpose - The co-word analysis method is commonly used to cluster-related keywords into the same keyword domain. In other words, traditional co-word analysis cannot cluster the same keywords into more than one keyword domain, and disregards the multi-domain property of keywords. The purpose of this paper is to propose an innovative keyword co-citation approach called "Complete Keyword Pair (CKP) method", which groups complete keyword sets of reference papers into clusters, and thus finds keywords belonging to more than one keyword domain, namely bridge-keywords. Design/methodology/approach - The approach regards complete author keywords of a paper as a complete keyword set to compute the relations among keywords. Any two complete keyword sets whose corresponding papers are co-referenced by the same paper are recorded as a CKP. A clustering method is performed with the correlation matrix computed from the frequency counts of the CKPs, for clustering the complete keyword sets. Since keywords may be involved in more than one complete keyword set, the same keywords may end up appearing in different clusters. Findings - Results of this study show that the CKP method can discover bridge-keywords with average precision of 80 per cent in the Journal of the Association for Computing Machinery citation bank during 2000-2006 when compared against the benchmark of Association for Computing Machinery Computing Classification System. Originality/value - Traditional co-word analysis focuses on co-occurrence of keywords, and therefore, cannot cluster the same keywords into more than one keyword domain. The CKP approach considers complete author keyword sets of reference papers to discover bridge-keywords. Therefore, the keyword recommendation system based on CKP can recommend keywords across multiple keyword domains via the bridge-keywords.