Noise analysis of an electro-optic sensor system

Tominaga, Mai; Okajima, Mei; Yamagishi, Mayuko; Shinagawa, Morikazu; Katsuyama, Jun; Matsumoto, Yoshinori; Tomosada, Nobuhiro

doi:10.1007/s10043-021-00701-4

Cited by 2 publications

(2 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Tominaga and others believe that too large feature space will make the evaluation of sample statistical characteristics more difficult, thus reducing the generalization ability of the classifier and causing the phenomenon of "over learning." Moreover, the processing of this high-dimensional vector has extremely high computational complexity, especially the so-called "dimension disaster" problem [12]. erefore, Yue and Wang said that how to retain those features that play an important role in classification and remove redundant features in order to reduce the total number of features, that is, how to carry out dimension reduction, has become an increasingly important research field [13].…”

Section: Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

English Text Analysis System Based on Genetic Algorithm

Yao

2022

Mobile Information Systems

View full text Add to dashboard Cite

In order to accurately extract the useful information in English, this paper studies English text analysis combined with a genetic algorithm and establishes a text analysis system. In this method, a text tendency analysis algorithm based on a genetic algorithm language model is proposed, and a Doc2vec text feature representation algorithm integrating the LDA model is designed; the parallelization technology of text analysis algorithm is studied, and the parallelization model of the algorithm by using spark big data platform is designed; the process of English text tendency analysis is studied, and a Chinese text analysis system is designed and implemented based on big data platform, including corpus intake, corpus annotation, corpus storage, model training, model verification, and other modules. In order to verify the feasibility of this subject, the accuracy of the Doc2vec text feature representation algorithm of the fused LDA model designed in the prototype system is tested. The experimental results show that the fused text representation model has high recognition degree, and the AUC value of the ROC curve reaches 0.95. At the same time, this paper tests the text analysis-related algorithms involved in the system. The test results show that the parallel algorithm can greatly improve the efficiency of the system.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

“…In formula (12), by calculating the standard deviation of the mutual information weight between the feature word and various texts, the dispersion degree of the weight value of the feature word and each category of information is expressed.…”

Section: Research and Analysis Of Text Feature Dimensionalitymentioning

confidence: 99%