• Journal of Information Processing Systems
• /
• v.17 no.4
• /
• pp.818-833
• /
• 2021

To address the problems of low precision rate, insufficient feature extraction, and poor contextual ability in existing text sentiment analysis methods, a mixed model account of a CNN-BiLSTM-TE (convolutional neural network, bidirectional long short-term memory, and topic extraction) model was proposed. First, Chinese text data was converted into vectors through the method of transfer learning by Word2Vec. Second, local features were extracted by the CNN model. Then, contextual information was extracted by the BiLSTM neural network and the emotional tendency was obtained using softmax. Finally, topics were extracted by the term frequency-inverse document frequency and K-means. Compared with the CNN, BiLSTM, and gate recurrent unit (GRU) models, the CNN-BiLSTM-TE model's F1-score was higher than other models by 0.0147, 0.006, and 0.0052, respectively. Then compared with CNN-LSTM, LSTM-CNN, and BiLSTM-CNN models, the F1-score was higher by 0.0071, 0.0038, and 0.0049, respectively. Experimental results showed that the CNN-BiLSTM-TE model can effectively improve various indicators in application. Lastly, performed scalability verification through a takeaway dataset, which has great value in practical applications.

• Wind and Structures
• /
• v.27 no.6
• /
• pp.357-368
• /
• 2018

For a long cable-stayed bridge, stay cables are its most important load-carrying components. In this paper, long-term monitoring of super-long stay cables of Sutong Bridge is introduced. A comprehensive data analysis procedure is presented, in which time domain and frequency domain based analyses are carried out. In time domain, the vibration data of several long stay cables are firstly analyzed and the standard deviation of the acceleration of stay cables, and its variation with time are obtained, as well as the relationship between in-plane vibration and out-plane vibration. Meanwhile, some vibrations such as wind and rain induced vibration are detected. Through frequency domain analysis, the basic frequencies of the stay cables are identified. Furthermore, the axial forces and their statistical parameters are acquired. To investigate the vibration deflection, an FFT-based decomposition method is used to get the modal deflection. In the end, the relationship between the vibration amplitude of stay cables and the wind speed is investigated based on correlation analysis. Through the adopted procedure, some structural parameters of the stay cables have been derived, which can be used for evaluating the component performance and corresponding management of stay cables.

• International Journal of Knowledge Content Development & Technology
• /
• v.9 no.3
• /
• pp.23-41
• /
• 2019

This study combines the concept of degree centrality in complex network with the Term Frequency $^*$ Proportional Document Frequency ($TF^*PDF$) algorithm; the combined method, called PMCN (PDF-Modified similarity and Complex Network), constructs relationship networks among sentences for writing news summaries. The PMCN method is a multi-document summarization extension of the ideas of Bun and Ishizuka (2002), who first published the $TF^*PDF$ algorithm for detecting hot topics. In their $TF^*PDF$ algorithm, Bun and Ishizuka defined the publisher of a news item as its channel. If the PDF weight of a term is higher than the weights of other terms, then the term is hotter than the other terms. However, this study attempts to develop summaries for news items. Because the $TF^*PDF$ algorithm summarizes daily news, PMCN replaces the concept of "channel" with "the date of the news event", and uses the resulting chronicle ordering for a multi-document summarization algorithm, of which the F-measure scores were 0.042 and 0.051 higher than LexRank for the famous d30001t and d30003t tasks, respectively.

• Journal of Home Health Care Nursing
• /
• v.26 no.2
• /
• pp.166-179
• /
• 2019

Purpose: The purpose of this study was to identify the job performance and educational needs, and recognition of professionalism among care workers, and to organize educational programs according to the priorities of care workers. Methods: The participants were 119 care workers who were working in long-term care facilities. Data were collected from May 31 to June 7, 2019 using self-report questionnaires. Collected data were analyzed using t-tests, ANOVA, and Spearman's Correlation Coefficients. Results: The performance aspects of the job were as follows: care for safety and infection-related, communication and leisure support, and excretion. The most demanded educational needs were in first-aid. Care workers had more than average professional recognition. Job performance and educational needs, and recognition of professionalism differed significantly according to several general characteristics. Conclusions: The educational needs of the areas with low frequency of job performance were high. First-aid is low in frequency, but it is important to cope with emergencies, so it is necessary to continue education. Also, there is a difference in recognition of professionalism according to the career. It will be necessary to develop individualized education programs to meet the needs of care workers.

• Korean Journal of Schizophrenia Research
• /
• v.23 no.2
• /
• pp.58-64
• /
• 2020

Objectives: In this study, we conducted an exploratory analysis of the current media trends on schizophrenia using text-mining methods. Methods: First, web-crawling techniques extracted text data from 575 news articles in 10 major newspapers between 2018 and 2019, which were selected by searching "schizophrenia" in the Naver News. We had developed document-term matrix (DTM) and/or term-document matrix (TDM) through pre-processing techniques. Through the use of DTM and TDM, frequency analysis, co-occurrence network analysis, and topic model analysis were conducted. Results: Frequency analysis showed that keywords such as "police," "mental illness," "admission," "patient," "crime," "apartment," "lethal weapon," "treatment," "Jinju," and "residents" were frequently mentioned in news articles on schizophrenia. Within the article text, many of these keywords were highly correlated with the term "schizophrenia" and were also interconnected with each other in the co-occurrence network. The latent Dirichlet allocation model presented 10 topics comprising a combination of keywords: "police-Jinju," "hospital-admission," "research-finding," "care-center," "schizophrenia-symptom," "society-issue," "family-mind," "woman-school," and "disabled-facilities." Conclusion: The results of the present study highlight that in recent years, the media has been reporting violence in patients with schizophrenia, thereby raising an important issue of hospitalization and community management of patients with schizophrenia.

• Phonetics and Speech Sciences
• /
• v.2 no.2
• /
• pp.37-42
• /
• 2010

The purpose of this study is to first discriminate and assess those infants who appear healthy in appearance but who could face possible risk factors in the future and, secondly, to identify those infants who may have difficulties in their developmental stages. The subjects of this study consisted of 35 full-term infants (39-40 weeks) and 33 pre-term infants (34-35 weeks). The infants' voices were recorded for three minutes, for which EDIROL by Roland and a stand-type microphone made by SONY were used. This was done to discern the value of the Breath unit (B-unit) and the fundamental frequencies ($F_0$). It was found that there were significant differences in terms of F0 since the pre-term infants had higher F0 than the full-term infants, showing a result of 436.4 Hz for the full-term infants and 460 Hz for the pre-term infants (p<.05) There was an average rate of 4.01 for the full-term infants and 4.02 (SD=1.69) for the pre-term infants in shimmer. For NHR, it was observed .44 for the full-term infants and .50 for the pre-term infants, thus revealing no significant differences in these observations. This study shows that the crying of newborn babies is related to their physical conditions and it is a sensatory response to these conditions. Furthermore, this study could be helpful for the early detection and measurement of newborn babies who look clinically healthy but could be at risk through acoustic and physiological analyses.

• Korean Journal of Clinical Laboratory Science
• /
• v.53 no.4
• /
• pp.333-341
• /
• 2021

Benzodiazepines (BDZs) drugs act on the GABA_A receptor, function as nerve suppressors, and are used to treat anxiety, insomnia, and panic disorder. We analyzed the data of 30 individuals to determine any differences in the sleep-electroencephalogram findings among individuals varying in age, benzodiazepine use, and duration of benzodiazepine use. Comparisons between users and non-users of benzodiazepines, short-term and long-term users, older and younger users, and older short-term and older long-term users, were achieved using electroencephalographic findings obtained through polysomnography. The parameters evaluated included sleep latency, sleep efficiency, sleep-stage percentages, number of sleep spindles, and average frequency of sleep-spindle. The difference between benzodiazepine users and non-users was significant with respect to sleep-stage percentages and average frequency of sleep-spindle. Older and younger users differed significantly with respect to sleep efficiency and sleep-stage percentages, whereas significant difference for sleep efficiency was obtained between long-term and short-term users. Taken together, our results indicate that BDZ consumption suppresses slow-wave sleep and increases the frequency of sleep spindles.

• Journal of Society of Occupational Therapy for the Aged and Dementia
• /
• v.12 no.2
• /
• pp.67-74
• /
• 2018

Objective : The purpose of this study is to quantitatively analyze the role of occupational therapy in long - term care insurance for the elderly using text mining, one of the big data analysis techniques. Method : For the analysis of newspaper articles, "Long - Term Care Insurance for the Elderly + Occupational Therapy for the Elderly" was collected after the period from 2007 to 208. Naver, which has a high share of the domestic search engine, utilized the database of Naver News by utilizing Textom, a web crawling tool. After collecting the article title and original text of 510 news data from the collection of the elderly long term care insurance + occupational therapy search, we analyzed the article frequency and key words by year. Result : In terms of the frequency of articles published by year, the number of articles published in 2015 and 2017 was the highest with 70 articles (13.7%), and the top 10 terms of the key word analysis showed the highest frequency of 'dementia' (344) In terms of key words, dementia, treatment, hospital, health, service, rehabilitation, facilities, institution, grade, elderly, professional, salary, industrial complex and people are related. Conclusion : In this study, it is meaningful that the textual mining technique was used to more objectively confirm the social needs and the role of the occupational therapist for the dementia and rehabilitation in the related key keywords based on the media reporting trend of the elderly long - term care insurance for 11 years. Based on the results of this study, future research should expand research field and period and supplement the research methodology through various analysis methods according to the year.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10b
• /
• pp.133-135
• /
• 2001

본 논문에서는 일반적으로 Word Based Matching 방법에서 많이 쓰이는 TFIDF 방법대신에 TTF(Total Term Frequency)와 ITTF(Inverse Total Term Frequecy) 에 가중치를 주어 문서분류의 정확도를 높이는 방법을 제안하고자 한다. TFIDF방법에서 IDF는 역문헌빈도를 나타내는데 Term에 대한 빈도비율의 공정성이 떨어져 문서 분류의 정확도에 한계가 있다. 본 논문에서 제시하는 문서 분류방법은 TTF와 ITTF에 각각의 가중치를 준 후에 차연산 이용하여 문서를 분류하는 것이다. 이러한 방법의 특징은 IDF를 사용할 때 보다 각 카테고리에 있는 term, 즉 단어의 중요도에 대한 가중치를 좀 더 공평하게 줌으로써 문서의 분류를 높일 수 있다. 본 논문에서는 조선일보의 카테고리를 사용하였으며 조선일보의 기사를 대상으로 문서 자동 분류 실험을 수행하였다. 실험 결과 TFIDF보다 본 논문에서 제안한 방법이 문서 분류에 높은 정확도를 나타냄을 보였다.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.67-67
• /
• 2000

In this paper, a hybrid type fuzzy controller is proposed to maintain molten steel level stable and reliable manner in high speed continuous casting regardless of various disturbances such as casting speed change, tundish weight variation, 치ogging/undoning of SEN(Submerged Entry Nozzle), periodic bulgings, etc. To accomplish this purpose, hardware filter and software filer are carefully designed to eliminate high frequency noise and to smooth input signals from harsh environments. In order to minimize the molten steel level variations from various disturbances the controller uses hybrid type control term: fuzzy logic term, proportional term, differential term and nonlinear feedback compensation tenn. The proposed controller is applied tn commercial mini-mill plant and shows considerable improvement in minimizing the molten steel variation.