• 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.

• Clinical and Experimental Reproductive Medicine
• /
• v.2 no.2
• /
• pp.27-36
• /
• 1975

The success of the family planning program in Korea is reflected in the decrease in the growth rate from 3.0% in 1962 to 2.0% in 1971, and in the decrease in the fertility rate from 43/1,000 in 1960 to 29/1,000 in 1970. However, it would be erroneous to attribute these reductions entirely to the family planning program. Other socio-economic factors, such as the increasing age at marriage and the increasing use of induced abortions, definitely had an impact on the lowered growth and fertility rate. Despite the relative success of the program to data in meeting its goals, there is no room for complacency. Meeting the goal of a further reduction in the population growth rate to 1.3% by 1981 is a much more difficult task than any one faced in the past. Not only must fertility be lowered further, but the size of the target population itself will expand tremendously in the late seventies; due to the post-war baby boom of the 1950's reaching reproductive ages. Furthermore, it is doubtful that the age at marriage will continue to rise as in the past or that the incidence of induced abortion will continue to increase. Consequently, future reductions in fertility will be more dependent on the performance of the national family planning program, with less assistance from these non-program factors. This paper will describe various approaches to help to the solution of these current problems. 1. PRACTICE RATE IN FAMILY PLANNING In 1973, the attitude (approval) and knowledge rates were quite high; 94% and 98% respectively. But a large gap exists between that and the actual practice rate, which is only 3695. Two factors must be considered in attempting to close the KAP-gap. The first is to change social norms, which still favor a larger family, increasing the practice rate cannot be done very quickly. The second point to consider is that the family planning program has not yet reached all the eligible women. A 1973 study determineded that a large portion, 3096 in fact, of all eligible women do not want more children, but are not practicing family planning. Thus, future efforts to help close the KAP-gap must focus attention and services on this important large group of potential acceptors. 2. CONTINUATION RATES Dissatisfaction with the loop and pill has resulted in high discontinuation rates. For example, a 1973 survey revealed that within the first six months initial loop acceptance. nearly 50% were dropouts, and that within the first four months of inital pill acceptance. nearly 50% were dropouts. These discontinuation rates have risen over the past few years. The high rate of discontinuance obviously decreases the contraceptive effectiveness. and has resulted in many unwanted births which is directly related to the increase of induced abortions. In the future, the family planning program must emphasize the improved quality of initial and follow-up services. rather than more quantity, in order to insure higher continuation rates and thus more effective contraceptive protection. 3. INDUCED ABORTION As noted earlier. the use of induced abortions has been increase yearly. For example, in 1960, the average number of abortions was 0.6 abortions per women in the 15-44 age range. By 1970. that had increased to 2 abortions per women. In 1966. 13% of all women between 15-44 had experienced at least one abortion. By 1971, that figure jumped to 28%. In 1973 alone, the total number of abortions was 400,000. Besides the ever incre.sing number of induced abortions, another change has that those who use abortions have shifted since 1965 to include- not. only the middle class, but also rural and low-income women. In the future. in response to the demand for abortion services among rural and low-income w~men, the government must provide and support abortion services for these women as a part of the national family planning program. 4. TARGET SYSTIi:M Since 1962, the nationwide target system has been used to set a target for each method, and the target number of acceptors is then apportioned out to various sub-areas according to the number of eligible couples in each area. Because these targets are set without consideration for demographic factors, particular tastes, prejudices, and previous patterns of acceptance in the area, a high discontinuation rate for all methods and a high wastage rate for the oral pill and condom results. In the future. to alleviate these problems of the methodbased target system. an alternative. such as the weighted-credit system, should be adopted on a nation wide basis. In this system. each contraceptive method is. assigned a specific number of points based upon the couple-years of protection (CYP) provided by the method. and no specific targets for each method are given. 5. INCREASE OF STERILIZA.TION TARGET Two special projects. the hospital-based family planning program and the armed forces program, has greatly contributed to the increasing acceptance in female and male sterilization respectively. From January-September 1974, 28,773 sterilizations were performed. During the same time in 1975, 46,894 were performed; a 63% increase. If this trend continues, by the end of 1975. approximately 70,000 sterilizations will have been performed. Sterilization is a much better method than both the loop and pill, in terms of more effective contraceptive protection and the almost zero dropout rate. In the future, the. family planning program should continue to stress the special programs which make more sterilizations possible. In particular, it should seek to add the laparoscope techniques to facilitate female sterilization acceptance rates. 6. INCREASE NUMBER OF PRIVATE ACCEPTORS Among the current family planning users, approximately 1/3 are in the private sector and thus do not- require government subsidy. The number of private acceptors increases with increasing urbanization and economic growth. To speed this process, the government initiated the special hospital based family planning program which is utilized mostly by the private sector. However, in the future, to further hasten the increase of private acceptors, the government should encourage doctors in private practice to provide family planning services, and provide the contraceptive supplies. This way, those do utilize the private medical system will also be able to receive family planning services and pay for it. Another means of increasing the number of private acceptors, IS to greatly expand the commercial outlets for pills and condoms beyond the existing service points of drugstores, hospitals, and health centers. 7. IE&C PROGRAM The current preferred family size is nearly twice as high as needed to achieve a stable poplation. Also, a strong boy preference hinders a small family size as nearly all couples fuel they must have at least one or more sons. The IE&C program must, in the future, strive to emphasize the values of the small family and equality of the sexes. A second problem for the IE&C program to work. with in the: future is the large group of people who approves family planning, want no more children, but do not practice. The IE&C program must work to motivate these people to accept family planning And finally, for those who already practice, an IE&C program in the future must stress continuation of use. The IE&C campaign, to insure highest effectiveness, should be based on a detailed factor analysis of contraceptive discontinuance. In conclusion, Korea faces a serious unfavorable sociodemographic situation- in the future unless the population growth rate can be curtailed. And in the future, the decrease in fertility will depend solely on the family planning program, as the effect of other socio-economic factors has already been maximumally felt. A second serious factor to consider is the increasing number of eligible women due to the 1950's baby boom. Thus, to meet these challenges, the program target must be increased and the program must improve the effectiveness of its current activities and develop new programs.

• Journal of Intelligence and Information Systems
• /
• v.27 no.3
• /
• pp.139-156
• /
• 2021

The biggest reason for using a deep learning model in image classification is that it is possible to consider the relationship between each region by extracting each region's features from the overall information of the image. However, the CNN model may not be suitable for emotional image data without the image's regional features. To solve the difficulty of classifying emotion images, many researchers each year propose a CNN-based architecture suitable for emotion images. Studies on the relationship between color and human emotion were also conducted, and results were derived that different emotions are induced according to color. In studies using deep learning, there have been studies that apply color information to image subtraction classification. The case where the image's color information is additionally used than the case where the classification model is trained with only the image improves the accuracy of classifying image emotions. This study proposes two ways to increase the accuracy by incorporating the result value after the model classifies an image's emotion. Both methods improve accuracy by modifying the result value based on statistics using the color of the picture. When performing the test by finding the two-color combinations most distributed for all training data, the two-color combinations most distributed for each test data image were found. The result values were corrected according to the color combination distribution. This method weights the result value obtained after the model classifies an image's emotion by creating an expression based on the log function and the exponential function. Emotion6, classified into six emotions, and Artphoto classified into eight categories were used for the image data. Densenet169, Mnasnet, Resnet101, Resnet152, and Vgg19 architectures were used for the CNN model, and the performance evaluation was compared before and after applying the two-stage learning to the CNN model. Inspired by color psychology, which deals with the relationship between colors and emotions, when creating a model that classifies an image's sentiment, we studied how to improve accuracy by modifying the result values based on color. Sixteen colors were used: red, orange, yellow, green, blue, indigo, purple, turquoise, pink, magenta, brown, gray, silver, gold, white, and black. It has meaning. Using Scikit-learn's Clustering, the seven colors that are primarily distributed in the image are checked. Then, the RGB coordinate values of the colors from the image are compared with the RGB coordinate values of the 16 colors presented in the above data. That is, it was converted to the closest color. Suppose three or more color combinations are selected. In that case, too many color combinations occur, resulting in a problem in which the distribution is scattered, so a situation fewer influences the result value. Therefore, to solve this problem, two-color combinations were found and weighted to the model. Before training, the most distributed color combinations were found for all training data images. The distribution of color combinations for each class was stored in a Python dictionary format to be used during testing. During the test, the two-color combinations that are most distributed for each test data image are found. After that, we checked how the color combinations were distributed in the training data and corrected the result. We devised several equations to weight the result value from the model based on the extracted color as described above. The data set was randomly divided by 80:20, and the model was verified using 20% of the data as a test set. After splitting the remaining 80% of the data into five divisions to perform 5-fold cross-validation, the model was trained five times using different verification datasets. Finally, the performance was checked using the test dataset that was previously separated. Adam was used as the activation function, and the learning rate was set to 0.01. The training was performed as much as 20 epochs, and if the validation loss value did not decrease during five epochs of learning, the experiment was stopped. Early tapping was set to load the model with the best validation loss value. The classification accuracy was better when the extracted information using color properties was used together than the case using only the CNN architecture.

• Korean Journal of Poultry Science
• /
• v.36 no.1
• /
• pp.39-45
• /
• 2009

This study was carried out to investigate the effect of dietary supplementation of Acanthopanax (A) senticosus and Eucommiaceae on the expression of lipogenic, myogenic and oxidative stress genes in broiler chickens. Birds were subjected (assigned) to one of the following 5 dietary treatments: control (CON), A. senticosus 0.5% (T1), 1.0% (T2), Eucommiaceae 0.5% (T3) and 1% (T4). Each treatment was replicated 8 times with 4 birds per replication, housed in 4 birds per cage. Birds were arranged according to randomized block design. Feeding trial was conducted from day 4 to 35th day of age. Liver and muscle tissues were collected for analysis. Broilers subjected to 1% A. senticosus had higher feed conversion ratio than the other treated birds whereas no significant differences were found in body weight, weight gain and feed intake. The gene expression levels of fatty acid synthase were not different among the treatments while the transcription factor $PPAR{\gamma}$ was highly expressed in Eucommiaceae but not in control and A. senticosus. The gene expression levels of myogenin were high in both A. senticosus and Eucommiaceae compared to control group. MyoD also showed high expression in treated groups furthermore, Eucommiaceae stimulated the expression of MyoD more than that of A. senticosus. The antioxidant gene expressions (SOD, CAT, SOD, GPX) generally were not much different among the treatments, however, SOD and GPX were stimulated in broilers consumed 1% Eucommiaceae diet. The result of this experiment showed that dietary supplementation of A. senticosus and Eucommiaceae in broiler may improve the antioxidant defence system through SOD and GPX without affect of growth performance in broilers.

• Communications of Mathematical Education
• /
• v.25 no.1
• /
• pp.99-118
• /
• 2011

For this case study of gifted education, two classrooms in two locations, show life in general of the gifted educational system. And for this case study the identity of teachers and the gifted, help to activate the mathematically gifted education for these research questions, which are as followed: Firstly, how is the gifted education classroom life? Secondly, what kind of identity do the teachers and gifted students bring to mathematics, mathematics teaching and mathematics learning? Being selected in the gifted children's education center solves the research problem of characteristic and approach. Backed by the condition and the permission possibility, 2 selected classes and 2 people, which are coming and going. Gifted education classroom life, the identity of teachers and gifted students in mathematics and mathematics teaching and mathematic learning. It will be for 3 months, with various recordings and vocal instruction between teacher and students. Collected observations and interviews will be analyzed over the course of instruction. The results analyzed include, social participation, structure, and the formation of the gifted education classroom life. The organization of classes were analyzed by the classes conscious levels to collect and retain data. The classes verification levels depended on the program's first class incentive, teaching and learning levels and understanding of gifted math. A performance assessment will be applied after the final lesson and a consultation with parents and students after the final class. The six kinds of social participation structure come out of the type of the most important roles in gifted education accounts, for these types of group discussions and interactions, students must have an interaction or individual activity that students can use, such as a work product through the real materials, which release teachers and other students for that type of questions to evaluate. In order for the development of meaningful mathematical concepts to formulate, mathematical principles require problem solving among all students, which will appear in the resolution or it will be impossible to map the meaning of the instruction from which it was formed. These results show the analysis of the mathematics, mathematics teaching, mathematics learning and about the identity of the teachers and gifted. Gifted education teachers are defined by gifted math, which is more difficult and requires more differentiated learning, suitable for gifted students. Gifted was defined when higher level math was created and challenged students to deeper thinking. Gifted students think that gifted math is creative learning and they are forward or passive to one-way according to the education atmosphere.

• Journal of Biosystems Engineering
• /
• v.5 no.1
• /
• pp.1-14
• /
• 1980

Threshing operation may be one of the most important processes in the paddy post-production system as far as the grain loss and labor requirement are concerned . head-feeding type threshers commercially available now in Korea originally were developed for threshing dry paddy in the range of 15 to 17 % in wet basis. However, threshing wet-paddy with the grain moisture content above 20 % has been strongly recommended, especially for new high-yielding Indica -type varieties ; (1) to reduce high grain loss incurred due to the handling operations, and (2) to prevent the quantitative and qualitative loss of milled -rice when unthreshed grains are rewetted due to the rainfall. The objective of this study were to investigate the adaptability of both a head-feeding type thresher and a throw-in type thresher to wet-paddy , and to find out the possiblilities of improving the components of these threshers threshing. Four varieties, Suweon 264 and Milyang 24 as Tongil sister line varieties, minehikari and Jinhueng as Japonica-type varieties, were used at the different levels of the moisture content of grains. Both the feed rate and the cylinder speed were varied for each material and each machine. The thresher output quality , composition of tailing return, and separating loss were analyzed from the sampels taken at each treatment. A separate experiment for measurement opf the power requirement of the head-feeding type thresher was also performed. The results are summarized as follows : 1. There was a difference in the thresher output quality between rice varieties. In case of wet-paddy threshing at 550 rpm , grains with branchlet and torn heads for the Suweon 264 were 12 % and 7 % of the total output in weight, respectively, and for the Minehikari 4.5 % and 2 % respectively. In case of dry paddy threshing , those for the Suweon 264 were 8 % and 5% , and for the Minehikari 4% and 1% respectively. However, those for the Milyang 23 , which is highly susceptable to shattering, were much lower with 1 % and 0.5% respectively, regardless of the moisture content of the paddy. Therefore, it is desirable to breed rice varieties of the same physical properties as well as to improve a thresher adaptable to all the varieties. Torn heads, which increased with the moisture content of rall the varieties except the Milyang 23 , decreased as the cylinder speed increased, but grains with branchlet didnt decrease. The damaged kernels increased with the cylinder speed. 3. The thresher output quality was not affected much by the feed rate. But grains with branchlet and torn heads increased slightly with the feed rate for the head-feeding type thresher since higher resistance lowered at the cylinder speed. 4. In order to reduce grains with branchlet and torn heads in wet-paddy threshing , it is desirable to improve the head-feeding type thresher by developing a new type of cylinder which to not give excess impact on kernels or a concave which has differenct sizes of holes at different locations along the cylinder. 5. For the head-feeding type thresher, there was a difference in separating loss between the varieties. At the cylinder speed of 600 rpm the separating losses for the Minehikari and the Suweon 264 were 1.2% and 0.6% respectively. The separating loss of the head-feeding type thresher was not affected by the moisture content of paddy while that of the Mini-aged thresher increased with the moisture content. 6. From the analysis of the tailings return , to appeared that the tailings return mechanism didn't function properly because lots of single grains and rubbishes were unnecessarily returned. 7. Adding a vibrating sieve to the head-feeding type thresher could increase the efficiency of separation. Consequently , the tailing return mechanism would function properly since unnecessary return could be educed greatly. 8. The power required for the head-feeding type thresher was not affected by the moisture content of paddy, but the average power increased linearly with the feed rate. The power also increased with the cylinder speed.

• Journal of Korean Society of Environmental Engineers
• /
• v.33 no.3
• /
• pp.212-221
• /
• 2011

Wastewater treatment plants (WWTPs) are required to meet the reinforced discharge standards which are differentiated as 0.2, 0.3 and 0.5 mg-TP/L for the district I, II and III, respectively. Although most of WWTPs are operating advanced biological phosphorus removal system, the supplementary phosphorus treatment facility using chemical addition should be required almost at all WWTPs. Therefore, water quality data from several exemplary full-scale plants operating phosphorus treatment process were analyzed to evaluate the reliability of removal performance. Additionally, a series of jar tests were conducted to find optimal coagulants dose for phosphorus removal by chemical precipitation and to describe characteristics of the reaction and sludge production. Chemical costs and the increasing sludge volume in physicochemical phosphorus removal process were estimated based on the results of jar tests. The minimum coagulant (aluminium sulfate and poly aluminium chloride) doses to keep TP concentration below 0.5 and 0.2 mg/L were around 25 and 30 mg/L (as $Al_2O_3$), respectively, in the mixed liquor of activated sludge. In the tertiary treatment facility, relatively lower coagulant doses of 1/12~1/3 the minimum doses for activated sludge were required to achieve the same TP concentrations of 0.2~0.5 mg/L. Increase in suspended solids concentration due to chemical precipitates in mixed liquor was estimated at 10~11%, compared to the concentration without chemical addition. When coagulant was added into mixed liquor, chemical (aluminium sulfate) cost was estimated to be 4~10 times higher than in secondary effluent coagulation/separation process. Sludge production to be wasted was also 4~10 times higher than secondary effluent coagulation/separation process.

• The Journal of Korean Orthopaedic Ultrasound Society
• /
• v.5 no.2
• /
• pp.66-74
• /
• 2012

Purpose: To compare the outcome of two methods of chronic calcific tendinitis (CCT) treatment, Multiple drilling alone versus combined drilling and extracorporeal shock-wave therapy (ESWT). Furthermore, to analyze the clinical and radiologic results of different energy level configurations of ESWT. Materials and Methods: Among the patients complaining shoulder pain who visited the clinic from June 2010 to August 2011, 98 were diagnosed with CCT of the supraspinatus and were divided into the following three groups. Multiple drilling alone (n=31), Multiple drilling followed by high-energy ESWT (n=31), Multiple drilling followed by low-energy ESWT (n=36). The study was conducted only with patients with chronic pain persisting longer than six months despite prolonged conservative therapy. Clinical evaluation was done before and after 12 weeks from treatment, in clinical terms using the ASES, KSS, CSS system reflecting performance and symptom improvement, and in radiologic terms by studying the change in size of the calcific nodules. Results: All of three groups showed effects for improvement of clinical function and decrease of calcification and clinical improvement was significantly high in comparison between the group fulfilled by only multiple needling (the third group) and the group fulfilled by additional ESWT (the first and second groups) and in the radiological evaluation, calcification size and the rate of calcification decrease showed significant improvement statistically. For the comparison among the groups, degree of clinical function improvement and rate of calcification decrease showed significant difference between high energy group (the first group) and multiple needling group (the third group) as well as low energy group (the second group) and multiple needling group (the third group). But, in comparison between high energy group (the first group) and low energy group (the second group), there was no significant difference for the degree of clinical function improvement and rate of calcification decrease. Conclusion: For the treatment of chronic calcific tendinitis, additional ESWT showed more superior effects on clinical function improvement and radiological improvement regardless of the energy standard rather than the exclusive fulfillment of needling. But, as the result of ESWT by the energy standard, there was no significant difference for the decrease of calcification and degree of clinical function improvement.

• Journal of Intelligence and Information Systems
• /
• v.24 no.2
• /
• pp.243-264
• /
• 2018

Although the worldwide battery market is recently spurring the development of lithium secondary battery, lead acid batteries (rechargeable batteries) which have good-performance and can be reused are consumed in a wide range of industry fields. However, lead-acid batteries have a serious problem in that deterioration of a battery makes progress quickly in the presence of that degradation of only one cell among several cells which is packed in a battery begins. To overcome this problem, previous researches have attempted to identify the mechanism of deterioration of a battery in many ways. However, most of previous researches have used data obtained in a laboratory to analyze the mechanism of deterioration of a battery but not used data obtained in a real world. The usage of real data can increase the feasibility and the applicability of the findings of a research. Therefore, this study aims to develop a model which predicts the battery deterioration using data obtained in real world. To this end, we collected data which presents change of battery state by attaching sensors enabling to monitor the battery condition in real time to dozens of golf carts operated in the real golf field. As a result, total 16,883 samples were obtained. And then, we developed a model which predicts a precursor phenomenon representing deterioration of a battery by analyzing the data collected from the sensors using machine learning techniques. As initial independent variables, we used 1) inbound time of a cart, 2) outbound time of a cart, 3) duration(from outbound time to charge time), 4) charge amount, 5) used amount, 6) charge efficiency, 7) lowest temperature of battery cell 1 to 6, 8) lowest voltage of battery cell 1 to 6, 9) highest voltage of battery cell 1 to 6, 10) voltage of battery cell 1 to 6 at the beginning of operation, 11) voltage of battery cell 1 to 6 at the end of charge, 12) used amount of battery cell 1 to 6 during operation, 13) used amount of battery during operation(Max-Min), 14) duration of battery use, and 15) highest current during operation. Since the values of the independent variables, lowest temperature of battery cell 1 to 6, lowest voltage of battery cell 1 to 6, highest voltage of battery cell 1 to 6, voltage of battery cell 1 to 6 at the beginning of operation, voltage of battery cell 1 to 6 at the end of charge, and used amount of battery cell 1 to 6 during operation are similar to that of each battery cell, we conducted principal component analysis using verimax orthogonal rotation in order to mitigate the multiple collinearity problem. According to the results, we made new variables by averaging the values of independent variables clustered together, and used them as final independent variables instead of origin variables, thereby reducing the dimension. We used decision tree, logistic regression, Bayesian network as algorithms for building prediction models. And also, we built prediction models using the bagging of each of them, the boosting of each of them, and RandomForest. Experimental results show that the prediction model using the bagging of decision tree yields the best accuracy of 89.3923%. This study has some limitations in that the additional variables which affect the deterioration of battery such as weather (temperature, humidity) and driving habits, did not considered, therefore, we would like to consider the them in the future research. However, the battery deterioration prediction model proposed in the present study is expected to enable effective and efficient management of battery used in the real filed by dramatically and to reduce the cost caused by not detecting battery deterioration accordingly.

• Journal of Navigation and Port Research
• /
• v.42 no.4
• /
• pp.245-252
• /
• 2018

A method of transmitting an alert signal in case of man-overboard (MOB) systems in a small fishing vessel navigating within coastal area is being operated as VHF-DSC equipment via a distress alert button and V-P ass Equipment via alert button or beacon separation. However, a small fishing vessel with a couple of crews on board is an inappropriate way to alert a man-overboard condition. However, internationally, MOB equipment using VHF-DSC, AIS, and Bluetooth technologies is used to transmit alert signals directly to the mother ship and other radio stations. In order to analyze the performance and technology of the MOB equipment operating in foreign countries, it was confirmed that the alarm signal can be received within a maximum of one nautical mile when the MOB device is on the water surface. An MOB device that meets domestic conditions needs to send an alarm signal to a station within the VHF communication range. However, in order to reduce the false alert signal, it is most appropriate to operate the VHF-DSC radio equipment installed on the ship remotely. Analysis of various technologies connecting the MOB and the VHF-DSC revealed that the Bluetooth system has advantages such as device miniaturization. When an emergency signal is transmitted from the MOB device, it can be received by a dedicated receiver and recognized through an external input terminal of the VHF-DSC equipment generating its own alarm. If the emergency situation cannot be processed at the mother ship, a distress alert is sent to all radio stations via VHF-DSC for response under emergencies faced by small fishing vessels.