Feature Selection for Classification
Feature selection has been the focus of interest for quite some time and much work has been done. With the creation of huge databases and the consequent requirements for good machine learning techn...
an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org
Author
7 papers · 1997–2021
showing 1–7 of 7 RSS
Feature selection has been the focus of interest for quite some time and much work has been done. With the creation of huge databases and the consequent requirements for good machine learning techn...
Che-Wun Chiou, Liuh Chii Lin, F.H. Chou +1 more
A low-complexity array multiplier for GF(2m) fields with an irreducible trinomial Xm+Xn+1 is presented. The space complexity of the proposed multiplier is reduced from order O(m2) to O(m) compared with the Lee's array multiplier. The time complexity of the proposed multiplier is about half that of Lee's array multiplier.
Che Wun Chiou, Chiou‐Yng Lee, Jim-Min Lin +3 more
Finite field multiplication over GF (2^m) is one of the most important arithmetic operations for Elliptic Curve Cryptosystem (ECC). Polynomial basis multipliers over GF (2^m) are widely applied in ECC due to its regular, modular, easily expansible benefits and the high suitability for VLSI implementation. This study will present a novel digit-serial polynomial basis multiplier using Karatsuba algorithm representation. To achieve efficient architectures, our proposed digit-serial architecture is different from existing digit-serial polynomial basis multipliers that use cut-set algorithm. The proposed digit-serial polynomial basis multiplier saves 90% space complexity as compared to existing similar studies. Existing digit-serial polynomial basis multipliers employ one dimensional array of digit cells, but our proposed digit-serial polynomial basis multiplier uses only one digit cell.
Yuan Meng, Yi-Xuan Guo, Xue-Yi Liuh +2 more
The prevention and control of COVID-19 epidemic is a great challenge for human beings today. In the battle against COVID-19, the hierarchical treatment measures based on symptom classifications have proved to be a particularly effective way to deal with the large-scale epidemic in the absence of adequate medical resources. This paper deals with the epidemic dynamic analyses of the COVID-19 based on a modified SEIR model with different symptoms. First, by taking symptom classifications and hierarchical treatments of patients into account, a modified SEIR model is established. Then, the proposed differential equations model is solved by using Runge-Kutta methods, and the parameters herein are estimated by least square principle based on the data released by the National Health Commission. Simulation results of the model show that the introduction of symptom classifications in the SEIR…
Yun Chi Yeh, Liuh Chii Lin, Yuan Dong Chen
Based on existing analog to digital converters (ADC), this study proposes a simple and reliable controlled phase shift (CPS) method to improve the sampling resolution of a periodic signal. The proposed method uses phase shifted sampling sets of a sequence of sampled periods to increase sample resolution. When combining these sets, the period is virtually sampled with a higher number of samples. The lower sampling rate (per period) alleviates the requirements to be imposed on the conversion time of the ADC. Under the original conversion time of ADC, the proposed method can provide more sampled signals. Experimental result shows the proposed CPS method can improve the sampling resolution of periodic signals.
The natural fault-tolerant properties and regular structure of the Lee-Lu-Lee's array multiplier over GF(2(superscript m)) fields make it very attractive for VLSI implementation. However, the Lee-Lu-Lee's array multiplier is time-consuming while comparing with other existing array multipliers. Thus, we will present fast array multipliers with multiple speeds as comparing with the Lee-Lu-Lee's array multiplier.
This dissertation discusses the dynamic formation of cultural identity in the modern Chinese self through postcolonial and Lacanian psychoanalysis, utilizing the history and political situation of modern China as a platform for the author's inquiry. Defeated by Western and Japanese colonial powers, China was forced to break with its long traditions and move towards modernity. The loss of cultural continuity formed the drive for establishing the modern Chinese self, and resulted in a complex involving both desire and resistance towards the West. The author suggests the tension of this ambiguous cultural identity can be resolved by developing an ability to establish a double identity, beyond the Chinese/Western power struggle, and reconnecting with original Chinese culture, which over thousands of years has been repressed.