• The Korean Journal of Rehabilitation Nursing
• /
• v.16 no.2
• /
• pp.90-99
• /
• 2013

Purpose: This study aimed to examine influence of cognitive function, disease severity and disability on ability to perform activity of daily living (ADL) after ischemic stroke. Methods: A total of 88 patients with ischemic stroke were recruited from January 1, 2008 to December 31, 2012 and assessed with the standardized cognitive test battery and self-reports about disease severity, disability, and ADL. To analyze the data, ANOVA, Pearson correlation coefficients and multiple regression were conducted using SPSS/WIN program. Results: Significant correlations were found between ADL and visuospatial function, visual memory, executive function, and disability (r=.29~.38). Executive function and disability explained 17.3% of total variability in ADL performance after ischemic stroke. Conclusion: Executive function may be a promising target for cognitive rehabilitation after ischemic stroke. Thus, effective therapeutic interventions such as cognitive training for stroke patients should be considered to improve their ability to perform ADL.

• Journal of Korean Academic Society of Home Health Care Nursing
• /
• v.9 no.2
• /
• pp.114-128
• /
• 2002

In order to provide information for the establishment and maintenance of a rehabilitation day care center for stroke patients. this study is to assess needs for the rehabilitation day care center of the stroke patients and to identify the factors influencing the needs for the center. The data were collected face-to-face interview with 223 stroke patients. using a structured questionnaire. from September 24. 2001 to November 20. 2001. Major findings are as follows. 1. Most of the participants($94.6\%$) needed rehabilitation day care center for stroke patients. $95.5\%$ of participants were willing to use the rehabilitation day care center. 2. Also the score of the needs for the center's health services was $2.84\pm60$ out of 4.00. In regards to the sub-contents. while the physical exercise therapy showed the highest mark($3.54\pm71$) in the needs. the following marks showed physical therapy($3.48\pm79$), training for the memory. thinking and judgment($3.30\pm93$). training for ADL($3.09\pm99$). health education program($3.04\pm93$). In the meantime. the expected effects from the use of the center are $2.89\pm61$ out of 4 and its sub-contents showed that the center would promote their physical and mental well-being($3.30\pm74$) and the center would be more effective than in home care($3.12\pm70$). 3. Meanwhile. the desired frequency of use in the future and distance had significant interrelation with their families living together(p<.05). In addition those who paid to use it differentiated significantly according to their ages and the types of insurance they had(p<.05). 4. The needs in degrees of speech disorder therapy and hobbies & amusements. the patients with other disease had significantly higher degrees than those patients without it (p<.05). Also in regard to the need degrees for physical therapy. healthy education programs and individual counseling including their families. the degrees of the patients with speech disorders were significantly lower than those of the patients without the disorder (p<.05). On the other hand. the patients with speech disorders were significantly higher than those patients without it in the need degree of the speech disorder therapy (p=.000). And the needs in degree concerning about speech disorder therapy. physical exercise therapy. training for ADL. medicinal substances therapy and family education were negatively correlated with the ADL (r=-.236$\sim$.305, (p<.005). 5. Finally. the expected effect of using the rehabilitation day care center showed significant differences statistically according to whether or not they had other disease (p<.05). In conclusion. the study showed the stroke patients were willing to use the center and had a high requirements for it and they especially had relatively high need degrees for the physical exercise therapy. physical therapy. training for memory. thinking and judgment. and healthy education program. And significant factors for the use of the center were their ages. types of insurance. family cohabitation. complications and speech disorders. ADL and so forth. Accordingly. the rehabilitation day care center needs to be established for the stroke patients and the center should develop rehabilitation care programs. which are individual and special programs customized for each patient's characteristics and health conditions.

• Journal of Digital Contents Society
• /
• v.19 no.6
• /
• pp.1051-1060
• /
• 2018

This study aims to propose instituting the serious games of language-based cognitive enhancement for the BIF children. The program consists of 4 cognitive areas (perception, attention, working memory, knowledge inference) in 4 language dimensions (phoneme, word, sentence, discourse). 16 games of 4 areas/2 dimensions with 3 difficulty levels were implemented in a mobile station and pilot-tested to children including BIFs. The results from the pilot tests supported for the validity and effectiveness of our games: Children's game performance correlated with their IQ scores (overall and sub-areas) revealing significant differences between the groups. The stroop scores in pre-and-post training hinted the increase of children's cognitive control.

• Journal of Advanced Navigation Technology
• /
• v.17 no.6
• /
• pp.726-735
• /
• 2013

This paper proposed face recognition methods about performance improvement of the face recognition using the properties of wavelet transform. Using discrete wavelet transform is Daubechies D4 filter that is similar to mother wavelet transform. For discrete wavelet transform method, In this case, by using LL subband only we can reduce processing time and amount of memory in recognition processing. To improve recognition ratio without further loss of 2 dimensional data changing, We applies 2D LDA. We perform SVM training algorithm to the feature vector obtained by 2D LDA. Experiment is performed using ORL database set and Yale database set by Matlab program. Test result shows that proposed method is superior to existence methods in recognition rate and performance time.

• KNOM Review
• /
• v.23 no.2
• /
• pp.18-28
• /
• 2020

With the development of the virtual community, the benefits that IT technology provides to people in fields such as healthcare, industry, communication, and culture are increasing, and the quality of life is also improving. Accordingly, there are various malicious attacks targeting the developed network environment. Firewalls and intrusion detection systems exist to detect these attacks in advance, but there is a limit to detecting malicious attacks that are evolving day by day. In order to solve this problem, intrusion detection research using machine learning is being actively conducted, but false positives and false negatives are occurring due to imbalance of the learning dataset. In this paper, a Random Oversampling method is used to solve the unbalance problem of the UNSW-NB15 dataset used for network intrusion detection. And through experiments, we compared and analyzed the accuracy, precision, recall, F1-score, training and prediction time, and hardware resource consumption of the models. Based on this study using the Random Oversampling method, we develop a more efficient network intrusion detection model study using other methods and high-performance models that can solve the unbalanced data problem.

• The Korean Journal of Archival Studies
• /
• no.12
• /
• pp.137-178
• /
• 2005

The anticommunist guerrillas' activities that aretheconcern of this article took place largely in North Korea or behind the enemy-held lines. Verifying their history is accordingly difficult and requires careful attention, but despite their active operations the military as well as the scholarly community have been lax in studying them. The Korean War came to be perceived as a traditional, limited war with regular battles, so that the studies addressed mostly the regular operations, and guerrilla warfare is remembered as an almost 'exclusive property' of the communist invaders; a small wonder that the anticommunist guerrillas have not been studied much and the collection of materials neglected. Therefore, in contrast with the witness accounts concerning regular battles, witness resources were of a small volume about these "patriots without the service numbers." For the above reasons the guerrilla participants and their later-organized fellowships took to the task of leaving records and compiling the histories of their units. They became active preservers of history in order to inform later generations of their works and also to secure deserved benefits from the government, in a world where none recognized their achievements. For instance, 4th Donkey Unit published witness accounts in addition to a unit history, and left video-recordings of guerrilla witnesses before any institute systematized the oral history of the guerrillas. In the case of Kyulsa ("Resolved to Die") Guerrilla Unit, the unit history was 10 times revised and expanded upon for publication, contributing substantially to the recovery of anticommunist guerrilla history which had almost totally lacked documented resources. Now because the guerrilla-related witness accounts were produced through fellowship societies and not individually, it often took the form of 'collective memory.' As a result, though thousands of former guerrillas remain surviving, the scarcity of numerous versions of, or perspectives upon, an event renders difficult an objective approach to the historical truth. Even requests to verify the service of a guerrilla member or to apply for decoration or government benefits for those killed in action, the process is taken care of not at the hands of the first party but the veteran society, so that a variety of opinions are not available for consideration. Moreover, some accounts were taken by American military personnel, and since some historians, unaware of official documents or evaluation of achievements, tended to center the records around their own units and especially to exaggerate the units' performances, they often featured factual errors. Thefollowing is the means to utilize positively the aforementioned type of witness accounts in military history research. It involves the active use of military historical detachments (MHD). As in the examples of those dispatched by the American forces during the Korean War, experts should be dispatched during, and not just after, wartimes. By considering and investigating the differences among various perspectives on the same historical event, even without extra documented resources it is possibleto arrive at theerrors or questionable points of the oral accounts, supplementing the additional accounts. Therefore any time lapses between witness accounts must be kept in consideration. Moreover when the oral accounts come from a group such as participants in the same guerrilla unit or operation, a standardized list of items ought to be put to use. Education in oral history is necessary not just for the training of experts. In America wherethefield sees much activity, it is used not only in college or graduate programs but also in elementary and lifetime educational processes. In comparison in our nation, and especially in historical disciplines, methodological insistence upon documented evidences prevails in the main, and in the fields of nationalist movement or modern history, oral accounts do not receive adequate attention. Like ancient documents and monuments, oral history also needs to be made a regular part of diverse resource materials at our academic institutes for history. Courses in memory and history, such as those in American colleges, are available possibilities.

• Journal of Intelligence and Information Systems
• /
• v.23 no.1
• /
• pp.95-108
• /
• 2017

Recently, AlphaGo which is Bakuk (Go) artificial intelligence program by Google DeepMind, had a huge victory against Lee Sedol. Many people thought that machines would not be able to win a man in Go games because the number of paths to make a one move is more than the number of atoms in the universe unlike chess, but the result was the opposite to what people predicted. After the match, artificial intelligence technology was focused as a core technology of the fourth industrial revolution and attracted attentions from various application domains. Especially, deep learning technique have been attracted as a core artificial intelligence technology used in the AlphaGo algorithm. The deep learning technique is already being applied to many problems. Especially, it shows good performance in image recognition field. In addition, it shows good performance in high dimensional data area such as voice, image and natural language, which was difficult to get good performance using existing machine learning techniques. However, in contrast, it is difficult to find deep leaning researches on traditional business data and structured data analysis. In this study, we tried to find out whether the deep learning techniques have been studied so far can be used not only for the recognition of high dimensional data but also for the binary classification problem of traditional business data analysis such as customer churn analysis, marketing response prediction, and default prediction. And we compare the performance of the deep learning techniques with that of traditional artificial neural network models. The experimental data in the paper is the telemarketing response data of a bank in Portugal. It has input variables such as age, occupation, loan status, and the number of previous telemarketing and has a binary target variable that records whether the customer intends to open an account or not. In this study, to evaluate the possibility of utilization of deep learning algorithms and techniques in binary classification problem, we compared the performance of various models using CNN, LSTM algorithm and dropout, which are widely used algorithms and techniques in deep learning, with that of MLP models which is a traditional artificial neural network model. However, since all the network design alternatives can not be tested due to the nature of the artificial neural network, the experiment was conducted based on restricted settings on the number of hidden layers, the number of neurons in the hidden layer, the number of output data (filters), and the application conditions of the dropout technique. The F1 Score was used to evaluate the performance of models to show how well the models work to classify the interesting class instead of the overall accuracy. The detail methods for applying each deep learning technique in the experiment is as follows. The CNN algorithm is a method that reads adjacent values from a specific value and recognizes the features, but it does not matter how close the distance of each business data field is because each field is usually independent. In this experiment, we set the filter size of the CNN algorithm as the number of fields to learn the whole characteristics of the data at once, and added a hidden layer to make decision based on the additional features. For the model having two LSTM layers, the input direction of the second layer is put in reversed position with first layer in order to reduce the influence from the position of each field. In the case of the dropout technique, we set the neurons to disappear with a probability of 0.5 for each hidden layer. The experimental results show that the predicted model with the highest F1 score was the CNN model using the dropout technique, and the next best model was the MLP model with two hidden layers using the dropout technique. In this study, we were able to get some findings as the experiment had proceeded. First, models using dropout techniques have a slightly more conservative prediction than those without dropout techniques, and it generally shows better performance in classification. Second, CNN models show better classification performance than MLP models. This is interesting because it has shown good performance in binary classification problems which it rarely have been applied to, as well as in the fields where it's effectiveness has been proven. Third, the LSTM algorithm seems to be unsuitable for binary classification problems because the training time is too long compared to the performance improvement. From these results, we can confirm that some of the deep learning algorithms can be applied to solve business binary classification problems.