• Journal of Intelligence and Information Systems
• /
• v.26 no.4
• /
• pp.1-25
• /
• 2020

Sentiment Analysis (SA) is a Natural Language Processing (NLP) task that analyzes the sentiments consumers or the public feel about an arbitrary object from written texts. Furthermore, Aspect-Based Sentiment Analysis (ABSA) is a fine-grained analysis of the sentiments towards each aspect of an object. Since having a more practical value in terms of business, ABSA is drawing attention from both academic and industrial organizations. When there is a review that says "The restaurant is expensive but the food is really fantastic", for example, the general SA evaluates the overall sentiment towards the 'restaurant' as 'positive', while ABSA identifies the restaurant's aspect 'price' as 'negative' and 'food' aspect as 'positive'. Thus, ABSA enables a more specific and effective marketing strategy. In order to perform ABSA, it is necessary to identify what are the aspect terms or aspect categories included in the text, and judge the sentiments towards them. Accordingly, there exist four main areas in ABSA; aspect term extraction, aspect category detection, Aspect Term Sentiment Classification (ATSC), and Aspect Category Sentiment Classification (ACSC). It is usually conducted by extracting aspect terms and then performing ATSC to analyze sentiments for the given aspect terms, or by extracting aspect categories and then performing ACSC to analyze sentiments for the given aspect category. Here, an aspect category is expressed in one or more aspect terms, or indirectly inferred by other words. In the preceding example sentence, 'price' and 'food' are both aspect categories, and the aspect category 'food' is expressed by the aspect term 'food' included in the review. If the review sentence includes 'pasta', 'steak', or 'grilled chicken special', these can all be aspect terms for the aspect category 'food'. As such, an aspect category referred to by one or more specific aspect terms is called an explicit aspect. On the other hand, the aspect category like 'price', which does not have any specific aspect terms but can be indirectly guessed with an emotional word 'expensive,' is called an implicit aspect. So far, the 'aspect category' has been used to avoid confusion about 'aspect term'. From now on, we will consider 'aspect category' and 'aspect' as the same concept and use the word 'aspect' more for convenience. And one thing to note is that ATSC analyzes the sentiment towards given aspect terms, so it deals only with explicit aspects, and ACSC treats not only explicit aspects but also implicit aspects. This study seeks to find answers to the following issues ignored in the previous studies when applying the BERT pre-trained language model to ACSC and derives superior ACSC models. First, is it more effective to reflect the output vector of tokens for aspect categories than to use only the final output vector of [CLS] token as a classification vector? Second, is there any performance difference between QA (Question Answering) and NLI (Natural Language Inference) types in the sentence-pair configuration of input data? Third, is there any performance difference according to the order of sentence including aspect category in the QA or NLI type sentence-pair configuration of input data? To achieve these research objectives, we implemented 12 ACSC models and conducted experiments on 4 English benchmark datasets. As a result, ACSC models that provide performance beyond the existing studies without expanding the training dataset were derived. In addition, it was found that it is more effective to reflect the output vector of the aspect category token than to use only the output vector for the [CLS] token as a classification vector. It was also found that QA type input generally provides better performance than NLI, and the order of the sentence with the aspect category in QA type is irrelevant with performance. There may be some differences depending on the characteristics of the dataset, but when using NLI type sentence-pair input, placing the sentence containing the aspect category second seems to provide better performance. The new methodology for designing the ACSC model used in this study could be similarly applied to other studies such as ATSC.

• Journal of Intelligence and Information Systems
• /
• v.27 no.1
• /
• pp.23-46
• /
• 2021

Collaborative filtering, which is often used in personalization recommendations, is recognized as a very useful technique to find similar customers and recommend products to them based on their purchase history. However, the traditional collaborative filtering technique has raised the question of having difficulty calculating the similarity for new customers or products due to the method of calculating similaritiesbased on direct connections and common features among customers. For this reason, a hybrid technique was designed to use content-based filtering techniques together. On the one hand, efforts have been made to solve these problems by applying the structural characteristics of social networks. This applies a method of indirectly calculating similarities through their similar customers placed between them. This means creating a customer's network based on purchasing data and calculating the similarity between the two based on the features of the network that indirectly connects the two customers within this network. Such similarity can be used as a measure to predict whether the target customer accepts recommendations. The centrality metrics of networks can be utilized for the calculation of these similarities. Different centrality metrics have important implications in that they may have different effects on recommended performance. In this study, furthermore, the effect of these centrality metrics on the performance of recommendation may vary depending on recommender algorithms. In addition, recommendation techniques using network analysis can be expected to contribute to increasing recommendation performance even if they apply not only to new customers or products but also to entire customers or products. By considering a customer's purchase of an item as a link generated between the customer and the item on the network, the prediction of user acceptance of recommendation is solved as a prediction of whether a new link will be created between them. As the classification models fit the purpose of solving the binary problem of whether the link is engaged or not, decision tree, k-nearest neighbors (KNN), logistic regression, artificial neural network, and support vector machine (SVM) are selected in the research. The data for performance evaluation used order data collected from an online shopping mall over four years and two months. Among them, the previous three years and eight months constitute social networks composed of and the experiment was conducted by organizing the data collected into the social network. The next four months' records were used to train and evaluate recommender models. Experiments with the centrality metrics applied to each model show that the recommendation acceptance rates of the centrality metrics are different for each algorithm at a meaningful level. In this work, we analyzed only four commonly used centrality metrics: degree centrality, betweenness centrality, closeness centrality, and eigenvector centrality. Eigenvector centrality records the lowest performance in all models except support vector machines. Closeness centrality and betweenness centrality show similar performance across all models. Degree centrality ranking moderate across overall models while betweenness centrality always ranking higher than degree centrality. Finally, closeness centrality is characterized by distinct differences in performance according to the model. It ranks first in logistic regression, artificial neural network, and decision tree withnumerically high performance. However, it only records very low rankings in support vector machine and K-neighborhood with low-performance levels. As the experiment results reveal, in a classification model, network centrality metrics over a subnetwork that connects the two nodes can effectively predict the connectivity between two nodes in a social network. Furthermore, each metric has a different performance depending on the classification model type. This result implies that choosing appropriate metrics for each algorithm can lead to achieving higher recommendation performance. In general, betweenness centrality can guarantee a high level of performance in any model. It would be possible to consider the introduction of proximity centrality to obtain higher performance for certain models.

• Journal of Intelligence and Information Systems
• /
• v.27 no.4
• /
• pp.1-22
• /
• 2021

Aspect Based Sentiment Analysis (ABSA), which analyzes sentiment based on aspects that appear in the text, is drawing attention because it can be used in various business industries. ABSA is a study that analyzes sentiment by aspects for multiple aspects that a text has. It is being studied in various forms depending on the purpose, such as analyzing all targets or just aspects and sentiments. Here, the aspect refers to the property of a target, and the target refers to the text that causes the sentiment. For example, for restaurant reviews, you could set the aspect into food taste, food price, quality of service, mood of the restaurant, etc. Also, if there is a review that says, "The pasta was delicious, but the salad was not," the words "steak" and "salad," which are directly mentioned in the sentence, become the "target." So far, in ABSA, most studies have analyzed sentiment only based on aspects or targets. However, even with the same aspects or targets, sentiment analysis may be inaccurate. Instances would be when aspects or sentiment are divided or when sentiment exists without a target. For example, sentences like, "Pizza and the salad were good, but the steak was disappointing." Although the aspect of this sentence is limited to "food," conflicting sentiments coexist. In addition, in the case of sentences such as "Shrimp was delicious, but the price was extravagant," although the target here is "shrimp," there are opposite sentiments coexisting that are dependent on the aspect. Finally, in sentences like "The food arrived too late and is cold now." there is no target (NULL), but it transmits a negative sentiment toward the aspect "service." Like this, failure to consider both aspects and targets - when sentiment or aspect is divided or when sentiment exists without a target - creates a dual dependency problem. To address this problem, this research analyzes sentiment by considering both aspects and targets (Target-Aspect-Sentiment Detection, hereby TASD). This study detected the limitations of existing research in the field of TASD: local contexts are not fully captured, and the number of epochs and batch size dramatically lowers the F1-score. The current model excels in spotting overall context and relations between each word. However, it struggles with phrases in the local context and is relatively slow when learning. Therefore, this study tries to improve the model's performance. To achieve the objective of this research, we additionally used auxiliary loss in aspect-sentiment classification by constructing CNN(Convolutional Neural Network) layers parallel to existing models. If existing models have analyzed aspect-sentiment through BERT encoding, Pooler, and Linear layers, this research added CNN layer-adaptive average pooling to existing models, and learning was progressed by adding additional loss values for aspect-sentiment to existing loss. In other words, when learning, the auxiliary loss, computed through CNN layers, allowed the local context to be captured more fitted. After learning, the model is designed to do aspect-sentiment analysis through the existing method. To evaluate the performance of this model, two datasets, SemEval-2015 task 12 and SemEval-2016 task 5, were used and the f1-score increased compared to the existing models. When the batch was 8 and epoch was 5, the difference was largest between the F1-score of existing models and this study with 29 and 45, respectively. Even when batch and epoch were adjusted, the F1-scores were higher than the existing models. It can be said that even when the batch and epoch numbers were small, they can be learned effectively compared to the existing models. Therefore, it can be useful in situations where resources are limited. Through this study, aspect-based sentiments can be more accurately analyzed. Through various uses in business, such as development or establishing marketing strategies, both consumers and sellers will be able to make efficient decisions. In addition, it is believed that the model can be fully learned and utilized by small businesses, those that do not have much data, given that they use a pre-training model and recorded a relatively high F1-score even with limited resources.

• Journal of agricultural medicine and community health
• /
• v.49 no.3
• /
• pp.257-270
• /
• 2024

Objective: This study aimed to empirically evaluate the effectiveness of chronic disease management services utilizing ICT for patients with chronic illnesses. Methods: From May to December, 2023, 452 people who were diagnosed with hypertension and diabetes at 9 participating public health centers were provided with customized health care services for 24 weeks, and 15 performance indicators were analyzed to evaluate their effectiveness. Results: Health behavior indicators and health risk factors decreased before and after participation in the project, blood pressure control rate, hypertension and diabetes management rate, medication compliance, weight, BMI, BP, WC, FBG, and HDL-cholesterol improved(p<0.001). Service factors that influence the improvement of health behaviors included the number of activity monitor transmissions(p=0.049), confirmed concentrated consultations on physical activity(p=0.003) and nutrition(p=0.005), and the adherence to medication missions for hypertension(p=0.020). As for service factors influencing chronic disease management, the improvement in blood pressure regulation rate was due to the number of times the blood pressure monitor was linked(p=0.004), and the number of confirmed intensive consultations on physical activity(p=0.026), and nutrition(p=0.049); the improvement in hypertension control rate was due to the number of times the activity monitor and blood pressure monitor were linked(p<0.001), and the number of hypertension medication missions carried out (p=0.004); and the improvement in diabetes control rate was due to the number of times the blood pressure monitor(p=0.022) and blood sugar system were linked(p=0.017). Conclusion: Although this study has limitations as a comparative study before and after the service, it has proved that chronic disease management using ICT has a positive effect on improvement of health behavior indicator, reduction of health risk factors, hypertension, diabetes management index, weight, BMI, TG, BP, FBG improvement.

• The Korean Journal of Nuclear Medicine
• /
• v.38 no.3
• /
• pp.259-267
• /
• 2004

Purpose: NEMA NU2-2001 was proposed as a new standard for performance evaluation of whole body PET scanners. in this study, system performance of Siemens CTI ECAT EXACT 47 PET scanner including spatial resolution, sensitivity, scatter fraction, and count rate performance in 2D and 3D mode was evaluated using this new standard method. Methods: ECAT EXACT 47 is a BGO crystal based PET scanner and covers an axial field of view (FOV) of 16.2 cm. Retractable septa allow 2D and 3D data acquisition. All the PET data were acquired according to the NEMA NU2-2001 protocols (coincidence window: 12 ns, energy window: $250{\sim}650$ keV). For the spatial resolution measurement, F-18 point source was placed at the center of the axial FOV((a) x=0, and y=1, (b)x=0, and y=10, (c)x=70, and y=0cm) and a position one fourth of the axial FOV from the center ((a) x=0, and y=1, (b)x=0, and y=10, (c)x=10, and y=0cm). In this case, x and y are transaxial horizontal and vertical, and z is the scanner's axial direction. Images were reconstructed using FBP with ramp filter without any post processing. To measure the system sensitivity, NEMA sensitivity phantom filled with F-18 solution and surrounded by $1{\sim}5$ aluminum sleeves were scanned at the center of transaxial FOV and 10 cm offset from the center. Attenuation free values of sensitivity wire estimated by extrapolating data to the zero wall thickness. NEMA scatter phantom with length of 70 cm was filled with F-18 or C-11solution (2D: 2,900 MBq, 3D: 407 MBq), and coincidence count rates wire measured for 7 half-lives to obtain noise equivalent count rate (MECR) and scatter fraction. We confirmed that dead time loss of the last flame were below 1%. Scatter fraction was estimated by averaging the true to background (staffer+random) ratios of last 3 frames in which the fractions of random rate art negligibly small. Results: Axial and transverse resolutions at 1cm offset from the center were 0.62 and 0.66 cm (FBP in 2D and 3D), and 0.67 and 0.69 cm (FBP in 2D and 3D). Axial, transverse radial, and transverse tangential resolutions at 10cm offset from the center were 0.72 and 0.68 cm (FBP in 2D and 3D), 0.63 and 0.66 cm (FBP in 2D and 3D), and 0.72 and 0.66 cm (FBP in 2D and 3D). Sensitivity values were 708.6 (2D), 2931.3 (3D) counts/sec/MBq at the center and 728.7 (2D, 3398.2 (3D) counts/sec/MBq at 10 cm offset from the center. Scatter fractions were 0.19 (2D) and 0.49 (3D). Peak true count rate and NECR were 64.0 kcps at 40.1 kBq/mL and 49.6 kcps at 40.1 kBq/mL in 2D and 53.7 kcps at 4.76 kBq/mL and 26.4 kcps at 4.47 kBq/mL in 3D. Conclusion: Information about the performance of CTI ECAT EXACT 47 PET scanner reported in this study will be useful for the quantitative analysis of data and determination of optimal image acquisition protocols using this widely used scanner for clinical and research purposes.

• Korean Security Journal
• /
• no.1
• /
• pp.371-408
• /
• 1997

This article is the reserch on actual condition of crime of arson which occurs in Korea and its countermeasures. The the presented problem in this article are that (1) we have generally very low rate concern about the crime of arson contrary to realistic problems of rapid increase of crime of arson (2) as such criminal motives became so diverse as to the economic or criminal purpose unlike characteristic and mental deficiency of old days, and to countermeasure these problems effectively it presentation the necessity of systemantic research. Based on analysis of reality of arson, the tendency of this arson in Korea in the ratio of increase is said to be higher than those in violence crime or general fire rate. and further its rate is far more greater than those of the U.S.A. and Japan. Arson is considered to be a method of using fire as crime and in case of presently residence to be the abject, it is a public offense crime which aqccompany fatality in human life. This is the well It now fact to all of us. And further in order to presentation to the crime of arson, strictness of criminal law (criminal law No, 164 and 169, and fire protection law No. 110 and 111) and classification of arsonist as felony are institutionary reinforced to punish with certainty of possibility, Therefore, as tendency of arson has been increased compared to other nations, it is necessary to supplement strategical policy to bring out overall concerns of the seriousness of risk and damage of arson, which have been resulted from the lack of understanding. In characteristics analysis of crime of arson, (1) It is now reveald that, in the past such crime rate appeared far more within the boundary of town or city areas in the past, presently increased rate of arsons in rural areas are far more than in the town or small city areas, thereby showing characteristics of crime of arson extending nation wide. (2) general timetable of arson shows that night more than day time rate, and reveald that is trait behavior in secrecy.(3) arsonists are usually arrested at site or by victim or report of third person(82,9%).Investigation activities or self surrenders rate only 11.2%. The time span of arrest is normally the same day of arson and at times it takes more than one year to arrest. This reveals its necessity to prepare for long period of time for arrest, (4) age rate of arson is in their thirties mostly as compared to homicide, robbery and adultery, and considerable numbers of arsons are in old age of over fifties. It reveals age rate is increased (5) Over half of the arsonists are below the junior high school (6) the rate of convicts by thier records is based on first offenders primarily and secondly more than 4 time convicts. This apparently shows necessity of effective correctional education policy for their social assimilation together with re-investigation of human education at the primary and secondary education system in thier life. The examples of motivation for arosnits, such as personal animosity, fury, monetary swindle, luscious purpose and other aims of destroying of proof, and other social resistance, violence including ways of threatening, beside the motives of individual defects, are diverse and arsonic suicide and specifically suicidal accompany together keenly manifested. When we take this fact with the criminal theory, it really reveals arsons of crime are increasing and its casualities are serious and a point as a way of suicide is the anomie theory of Durkheim and comensurate with the theory of that of Merton, Specifically in the arson of industrial complex, it is revealed that one with revolutionary motive or revolting motive would do the arsonic act. For the policy of prevention of arsons, professional research work in organizational cooperation for preventive activities is conducted in municipal or city wise functions in the name of Parson Taskforces and beside a variety of research institutes in federal government have been operating effectively to countermeasure in many fields of research. Franch and Sweden beside the U.S. set up a overall operation of fire prevention research funtions and have obtained very successful result. Japan also put their research likewise for countermeasure. In this research as a way of preventive fire policy, first, it is necessary to accomodate the legal preventitive activities for fire prevention in judicial side and as an administrative side, (1) precise statistic management of crime of arson (2) establishment of professional research functions or a corporate (3) improvement of system for cooperative structural team for investigation of fires and menpower organization of professional members. Secondly, social mentality in individual prospect, recognition of fires by arson and youth education of such effect, educational program for development and practical promotion. Thirdly, in view of environmental side, the ways of actual performance by programming with the establishment of cooperative advancement in local social function elements with administrative office, habitants, school facilities and newspapers measures (2) establishment of personal protection where weak menpowers are displayed in special fire prevention measures. These measures are presented for prevention of crime of arson. The control of crime and prevention shall be prepared as a means of self defence by the principle of self responsibility Specifically arsonists usually aims at the comparatively weak control of fire prevention is prevalent and it is therefore necessary to prepare individual facilities with their spontaneous management of fire prevention instead of public municipal funtures of local geverment. As Clifford L. Karchmer asserted instead of concerns about who would commit arson, what portion of area would be the target of the arson. It is effective to minister spontaveously the fire prevention measure in his facility with the consideration of characteristics of arson. On the other hand, it is necessary for the concerned personnel of local goverment and groups to distribute to the local society in timely manner for new information about the fire prevention, thus contribute to effective result of fire prevention result. In consideration of these factors, it is inevitable to never let coincide with the phemonemon of arsons in similar or mimic features as recognized that these could prevail just an epedemic as a strong imitational attitude. In processing of policy to encounter these problems, it is necessary to place priority of city policy to enhancement of overall concerns toward the definitive essense of crime of arson.

• The Journal of Korean Society for Radiation Therapy
• /
• v.16 no.2
• /
• pp.9-17
• /
• 2004

Purpose : Although Improve of CT, MRI Radio-diagnosis and Radiation Therapy Planing, but we still use ICRU38 Planning system(2D film-based) broadly. 3-Dimensional ICR plan(CT image based) is not only offer tumor and normal tissue dose but also support DVH information. On this study, we plan irradiation-goal dose on CTV(CTV plan) and irradiation-goal dose on ICRU 38 point(ICRU38 plan) by use CT image. And compare with tumor-dose, rectal-dose, bladder-dose on both planning, and analysis DVH Method and Material : Sample 11 patients who treated by Ir-192 HDR. After 40Gy external radiation therapy, ICR plan established. All the patients carry out CT-image scanned by CT-simulator. And we use PLATO(Nucletron) v.14.2 planing system. We draw CTV, rectum, bladder on the CT image. And establish plan irradiation-$100\%$ dose on CTV(CTV plan) and irradiation-$100\%$ dose on A-point(ICRU38 plan) Result : CTV volume($average{\pm}SD$) is $21.8{\pm}26.6cm^3$, rectum volume($average{\pm}SD$) is $60.9{\pm}25.0cm^3$, bladder volume($average{\pm}SD$) is $116.1{\pm}40.1cm^3$ sampled 11 patients. The volume including $100\%$ dose is $126.7{\pm}18.9cm^3$ on ICRU plan and $98.2{\pm}74.5cm^3$ on CTV plan. On ICRU planning, the other one's $22.0cm^3$ CTV volume who residual tumor size excess 4cm is not including $100\%$ isodose. 8 patient's $12.9{\pm}5.9cm^3$ tumor volume who residual tumor size belows 4cm irradiated $100\%$ dose. Bladder dose(recommended by ICRU 38) is $90.1{\pm}21.3\%$ on ICRU plan, $68.7{\pm}26.6\%$ on CTV plan, and rectal dose is $86.4{\pm}18.3\%,\;76.9{\pm}15.6\%$. Bladder and Rectum maximum dose is $137.2{\pm}50.1\%,\;101.1{\pm}41.8\%$ on ICRU plan, $107.6{\pm}47.9\%,\;86.9{\pm}30.8\%$ on CTV plan. Therefore CTV plan more less normal issue-irradiated dose than ICRU plan. But one patient case who residual tumor size excess 4cm, Normal tissue dose more higher than critical dose remarkably on CTV plan. $80\%$over-Irradiated rectal dose(V80rec) is $1.8{\pm}2.4cm^3$ on ICRU plan, $0.7{\pm}1.0cm^3$ on CTV plan. $80\%$over-Irradiated bladder dose(V80bla) is $12.2{\pm}8.9cm^3$ on ICRU plan, $3.5{\pm}4.1cm^3$ on CTV plan. Likewise, CTV plan more less irradiated normal tissue than ICRU38 plan. Conclusion : Although, prove effect and stability about previous ICRU plan, if we use CTV plan by CT image, we will reduce normal tissue dose and irradiated goal-dose at residual tumor on small residual tumor case. But bigger residual tumor case, we need more research about effective 3D-planning.

• (The)Study of the Eastern Classic
• /
• no.69
• /
• pp.83-114
• /
• 2017

The purpose of this article is to analyze the standings of working-level officials belonging to Jongbusi (Court of the Royal Clan) holding the rank of "jeong" and below between the 18th and mid-19th Century. Jongbusi, which was headed by a Grade-3 official, was in charge of the compilation of royal genealogy and supervision of royal relatives. During the late Joseon Period, its officials were composed of its chief, jeong, jubu (Grade-6 official), and jikjang (Grade-7 official). By 1864, it was incorporated into Jongchinbu (Office of the Royal Relatives). Jongbusi nangcheong seonsaengan (Register of Staff Officials at the Court of the Royal Clan), which is preserved at the Jangseogak Archives of the Academy of Korean Studies, lists the officials who served at the office between 1794 and its incorporation into Jongchinbu in 1864. The register also includes the officials' ranks, names, DOBs, family clans, their ranks in the offices they were transferred from, their ranks in the office they were transferred to, etc. Those interested view it as a precious relic that provides valuable information on the officialdom of the dynasty. A majority of the officials who served at Jongbusi were those who passed the higher civil service examination. Many of them at the level of jikjang were those who passed the licentiate examination. Their designation as an official was part of the "muneum" system, which granted official posts to descendants of those who accomplished a distinguished service for the country or served as a high-ranking official. They were those transferred from equal or lower positions in another office. Many of jubu-level officials of Jongbusi were those transferred from honorable and important posts of other offices or local administrative offices. Many of jikjang-level officials of Jongbusi were those who previously served as dosa (assistant officials) at Uigeumbu (Bureau of Crime Investigation) headed by a Grade-1 official. The officials' transfer to an office with a lower position like Jongbusi appears to have been for the provision of placing them in working-level positions rather than letting them remain in positions only carrying an honorary title. As for the transfer of officials of Jongbusi to other offices, many of those with the rank of jeong were transferred to lower positions. Supposedly it was because not many Grade-3 positions were vacant. Many of them were transferred to honorable and important posts. Some of them were also transferred to positions at local government offices, supposedly to avoid an excess of personnel at the central government. Those at the level of jubu or jikjang of Jongbusi were transferred to equal or higher posts in other offices. Particularly, most of those holding the position of jikjang (Grade-7) were transferred to higher posts. The family clan that produced the largest number (10%) of Jongbusi officials was the Jeonju Yi Clan, which produced the largest number of those who passed the higher civil service examination. It was also found that the top 20 family clans produced about half of the entirety of Jongbusi officials. According to the aforesaid Jongbusi nangcheong seonsaengan, about 90% of the cases of promotion of Jongbusi officials occurred after the revision of Seonwon boryak (Royal Genealogy of the Joseon Dynasty). It is speculated that the supervision of royal family members, one of the two leading functions assigned to Jongbusi, was suspended in the late Joseon Period. The relevant function does not appear even in chronicles pertaining to the Joseon Dynasty. The reason being had something to do with the sharp decrease in the number of royal family members during the reign of King Injo (r. 1623-1649). Their number was decreased to the extent that royal ceremonies could not be adequately carried out. Naturally, the meaning of supervising royal family members faded. Witnessing such a sorry state of the royal family, Heungseon Daewongun, King Gojong's father who served as the regent, incorporated Jongbusi into Jongchinbu in an effort to enhance the status and authority of the royal family.

• Journal of Intelligence and Information Systems
• /
• v.24 no.1
• /
• pp.205-225
• /
• 2018

Convolutional Neural Network (ConvNet) is one class of the powerful Deep Neural Network that can analyze and learn hierarchies of visual features. Originally, first neural network (Neocognitron) was introduced in the 80s. At that time, the neural network was not broadly used in both industry and academic field by cause of large-scale dataset shortage and low computational power. However, after a few decades later in 2012, Krizhevsky made a breakthrough on ILSVRC-12 visual recognition competition using Convolutional Neural Network. That breakthrough revived people interest in the neural network. The success of Convolutional Neural Network is achieved with two main factors. First of them is the emergence of advanced hardware (GPUs) for sufficient parallel computation. Second is the availability of large-scale datasets such as ImageNet (ILSVRC) dataset for training. Unfortunately, many new domains are bottlenecked by these factors. For most domains, it is difficult and requires lots of effort to gather large-scale dataset to train a ConvNet. Moreover, even if we have a large-scale dataset, training ConvNet from scratch is required expensive resource and time-consuming. These two obstacles can be solved by using transfer learning. Transfer learning is a method for transferring the knowledge from a source domain to new domain. There are two major Transfer learning cases. First one is ConvNet as fixed feature extractor, and the second one is Fine-tune the ConvNet on a new dataset. In the first case, using pre-trained ConvNet (such as on ImageNet) to compute feed-forward activations of the image into the ConvNet and extract activation features from specific layers. In the second case, replacing and retraining the ConvNet classifier on the new dataset, then fine-tune the weights of the pre-trained network with the backpropagation. In this paper, we focus on using multiple ConvNet layers as a fixed feature extractor only. However, applying features with high dimensional complexity that is directly extracted from multiple ConvNet layers is still a challenging problem. We observe that features extracted from multiple ConvNet layers address the different characteristics of the image which means better representation could be obtained by finding the optimal combination of multiple ConvNet layers. Based on that observation, we propose to employ multiple ConvNet layer representations for transfer learning instead of a single ConvNet layer representation. Overall, our primary pipeline has three steps. Firstly, images from target task are given as input to ConvNet, then that image will be feed-forwarded into pre-trained AlexNet, and the activation features from three fully connected convolutional layers are extracted. Secondly, activation features of three ConvNet layers are concatenated to obtain multiple ConvNet layers representation because it will gain more information about an image. When three fully connected layer features concatenated, the occurring image representation would have 9192 (4096+4096+1000) dimension features. However, features extracted from multiple ConvNet layers are redundant and noisy since they are extracted from the same ConvNet. Thus, a third step, we will use Principal Component Analysis (PCA) to select salient features before the training phase. When salient features are obtained, the classifier can classify image more accurately, and the performance of transfer learning can be improved. To evaluate proposed method, experiments are conducted in three standard datasets (Caltech-256, VOC07, and SUN397) to compare multiple ConvNet layer representations against single ConvNet layer representation by using PCA for feature selection and dimension reduction. Our experiments demonstrated the importance of feature selection for multiple ConvNet layer representation. Moreover, our proposed approach achieved 75.6% accuracy compared to 73.9% accuracy achieved by FC7 layer on the Caltech-256 dataset, 73.1% accuracy compared to 69.2% accuracy achieved by FC8 layer on the VOC07 dataset, 52.2% accuracy compared to 48.7% accuracy achieved by FC7 layer on the SUN397 dataset. We also showed that our proposed approach achieved superior performance, 2.8%, 2.1% and 3.1% accuracy improvement on Caltech-256, VOC07, and SUN397 dataset respectively compare to existing work.

• Journal of the Korean Home Economics Association
• /
• v.19 no.3
• /
• pp.83-101
• /
• 1981

Under the full consideration of the growing need and importance of population education in the field of home economics in Korea, the study was carried out to verify and assess the following facts on the current issues of population education of home economists who are presently engaging in teaching professions as the teachers of middle and high school and professors of college and universities by setting its primary objectives of the study as followings; 1) to assess the degree of general knowledge and attitudes of home economists toward population education in the field of home economics, 2) to verify the problems encountered in implementing population education by home economists in its field, 3) to find an existing status of previous trainings received and other activities of population education of home economists aimed at utilizing these findings as a part of reference materials when the population education is conducted in the field of home economics. In order to attain these objectives described above, the questionnaire was carefully designed to house a total of 40 questions with good combination of multiple-choice and the simple answer questions. The mail questionnaire survey was conducted by establishing teachers of home economics at middle/high schools and college/universities as Senior Home Economists(SHE) who are from public, private liberal arts and vocational schools. The rate of response observed during the survey was 45.6 percent and the findings of the survey research are as follows: 1) Examining the status of the respondents by residence and religion, it was found that 45 percent of middle & high school teachers ar.d 59. 1 percent of college professors are residing in Seoul city area and that the largest percent of them are christian in their religion. Analyzing respondents by their ages, 56 percent of middle/high school teachers are in their 30s, 45 percent of college professors are in their 40s, and 37 percent of college teachers are in their 30s. In addition, 13 percent of the total respondents are found to be unmarried. The study also revealed that 71 percent of the college professors finished Master Degree course and 82 percent of middle/high school teachers are graduated from college level lasting 4 years. Looking over the status cf major fields of respondents, 68.4 percent of middle/high school teachers are specialized in home economic education and the college professors, on the other hand, show relatively even prortion by specializing in the order of food & nutrition science, clothes & textile science and home managerial science. As far as the length of teaching experience is concerned, a relatively longer period of teaching experience is observed in the college professors in comparison with that of middle/high school teachers. In other words, 33.3 percent of middle/high school teachers are experienced in teaching from 6 to 10 years on average while 43.9 percent of college professors show more than 16 years of experience. 2) Examining the status of existing number of children cf the respondents, one boy and one daughter pattern is predominant, showing 28.5 percent in middle/high school teachers and 21.1 percent in college professors. As for the desired number of children of unmarried respondents, it is observed that 43.8 percent of middle/high school teachers desire to have one boy and one girl, and 31.3 percent of college professors want to have one child regardless of the sex. By assessing the degree of awareness of the population education through their students, it is observed that 53 percent of middle/high school teachers and 50 percent of college professors are aware of population education in some extent and that a majority of respondents took the positive attitudes toward an inclusion of family planning components into the formal school education. Another noteworthy to observe is that a total of 84.8 percent out of middle/high school teachers pointed that the population education currently conducted at schools as a part of home economics are less sufficient than it should be. 3) Analyzing the tendency as to whether the respondents were experienced in receiving population education during the time when they were students, 75 percent of college professors and 59 percent of middle/high school teachers responded negative answers in the survey. In the mean time, a total of 50 percent of the respondents replied that they began to acknowledge the importance of population education mainly through the participation of some sort of population-education orientend seminars, experienced by 40 percent of college professors and 80 percent of middle/high school teachers. 4) What it calls attention in this study was to find that 96.5 percent of middle/high school teachers and 72 percent of college professors conduct population education to some extent during their lecture hours and that more than 80 percent of them are never experienced in teaching population and family planning contents in their regular classes. It is, on the other hand, found that no more than once was the response of those who believe themselves that they are experienced in teaching these relevant components to their students. Analyzing the contents of the subjects being taught in the class, a large percent of them are found to be consisted of population and family planning contents. According to this study, the current population education through the formal school is quite inactive. Analyzing the facts, 44.9 percent of the college professors responded that the population and family planning components are quite apart from their specialization which eventually generates lack of interest in the field. 5) It is also noticed through the study that the degree of frequency of commenting on population and family planning contents during the classes was depending significantly on their specializations which means that the degree of frequency varies from a major to another. Those who majored in home managerial science was the first one, as compared to others who majored in different specializations. Glancing over the status of correlations between ages of the respondents and numbers of seminar paticipation, it is quite clear that the aged group participated more than the younger group did, and that the most highest number of participations made by college professors were those who are in 50s. In addition, it is also found that those who are aged 20s and 60s of the respondents were the group who comments least on the contents of population and family planning at their classes. The suggestions and recommendation made through this survey research are as follows. 1) No one denies that the rapid increase of population, as compared to the limited size of land and resources, will certainly affect adversly to an enhancement of individual life quality which will, eventually, bring forth the poverty of the nation. This is the reasson why we are insisting that the world population be controlled up to an optimum level with a matter of global concerns. It is our understading that the primary aim for reducing number of population is believed to be attained only by conducting the systematic and comprehensive population education through the formal schools. Therefore, the role of home economists in the field of population/family planning education is considered very importment due to the fact that an ultimate goal of population education is placed in elevating the quality of family life by having optimum number of children through family planning program. 2) It is quite clear that home economists as teachers of formal school in all level are invited to pay their attention on redefining the ultimate goal of education and that of population education. We also understant that the primary objective of population education is to change the norm and value of the clients by replenishing the students with pertinent knowledge and attitudes on population and its related problems through a sort of education in order to attain the ultimate goal for enhancing the quality of life. There is no exception in the theory of home economics. An altimate goal of home economics is to elevate the general quality of life through an establishment of value existed in daily life. Considering the relations between population education and home economics, it is quite indespensable to bandle population components as an integral part in the field of home economics. We believe, therefore, that the senior home economists positive participation in the effort population control is more needed than it has been. 3) It is also strongly urged that population education should be a part of instructor training course for home economics. In other words, the teacher of home economics should be well aware of population and its problems by teaching interrelationship between population education and home economics, needs, contents and methods of population education during the instructor training courese for home economics. In addition, the senior home economists should be encouraged through positive participation on the short term training by types of domestic and international seminar, workshop, etc. 4) We certainly believe that the population education can not sustain itself without any backing-up of information and findings' of various and comprehensive researches of natural and social sciences. Accordingly, every senior home economist is invited to exert their maximum effort to conduct systematic study with an aim to utilize these findings and information at best in population education in the field of home economics. Therefore, we consider that the development of training material is imminent in order to provide effective and efficient population education through the for training of home economies. It should be noted that these training materials must be carefully designed, tailored and developed to meet the different classes of trainees under the considerations as to whether it is easily adaptable and infusable into the curricula of every field of home economics, and it is acceptable in the degree of difficulty and quality in its contents. 5) It is true that there are many domestic and international research rapers, reports and findings in the field of population education and family planning. However, there is a tendency that the most of research papers are heavily relying on the authors intension and preferences in its expression and publication. Under these circumstances, it is urged that the home economists should aware of the growing need of the technical training in order to keep these available information and research findings reprocessed and redesigned to insure the practical application into the population education in the field of home economics in Korea.