• Journal of KIISE
• /
• v.43 no.2
• /
• pp.269-274
• /
• 2016

Named entity recognition is a process which extracts named entities in sentences and determines categories of the named entities. Previous studies on named entity recognition have primarily been used for supervised learning. For supervised learning, a large training corpus manually annotated with named entity categories is needed, and it is a time-consuming and labor-intensive job to manually construct a large training corpus. We propose a semi-supervised learning method to minimize the cost needed for training corpus construction and to rapidly enhance the performance of named entity recognition. The proposed method uses distance supervision for the construction of the initial training corpus. It can then effectively remove noise sentences in the initial training corpus through the use of an active bagging method, an ensemble method of bagging and active learning. In the experiments, the proposed method improved the F1-score of named entity recognition from 67.36% to 76.42% after active bagging for 15 times.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.11
• /
• pp.5427-5445
• /
• 2019

While deep neural networks have achieved remarkable performance in representation learning, a huge amount of labeled training data are usually required by supervised deep models such as convolutional neural networks. In this paper, we propose a new representation learning method, namely generative adversarial networks (GAN) based bagging deep convolutional autoencoders (GAN-BDCAE), which can map data to diverse hierarchical representations in an unsupervised fashion. To boost the size of training data, to train deep model and to aggregate diverse learning machines are the three principal avenues towards increasing the capabilities of representation learning of neural networks. We focus on combining those three techniques. To this aim, we adopt GAN for realistic unlabeled sample generation and bagging deep convolutional autoencoders (BDCAE) for robust feature learning. The proposed method improves the discriminative ability of learned feature embedding for solving subsequent pattern recognition problems. We evaluate our approach on three standard benchmarks and demonstrate the superiority of the proposed method compared to traditional unsupervised learning methods.

• Journal of Practical Agriculture & Fisheries Research
• /
• v.11 no.1
• /
• pp.79-94
• /
• 2009

The research project was conducted to develop a paper bagging machine for grape. This technology was aimed to highly reduce a labor for paper bagging in grape and bakery. In agriculture labor and farm population has rapidly decreased since 1980 in Korea so there was so limit in labor. In particular there is highly population in women and old age at rural area and thus labor cost is so high. Therefore a labor saving technology in agricultural sector might be needed to be replaced these old age with mechanical and labor saving tool in agriculture. The following was summarized of the research results for development of a paper bagging machine for grape. 1. Development of a new paper bagging machine for grape - This machine was designed by CATIA VI2/AUTO CAD2000 programme. - A paper bagging machine was mechanically binded a paper bag of grape which should be light and small size. This machine would be designed for women and old age with convenience during bagging work at the field site. - This machine was manufactured with total weight of less than 350g. - An overage bagging operation was more than 99% at the actual field process. - A paper bagging machine was designed with cartridge type which would be easily operated between rows and grape branches under field condition. - The type of cartridge pin was designed as a C-ring type with the length of 500mm which was good for bagging both grape and bakery. - In particular this machine was developed to easily operated among vines of the grape trees. 2. Field trials of a paper bagging machine in grape - There was high in grape quality as compared to the untreated control at the application of paper bagging machine. - The efficiency of paper bagging machine was 102% which was alternative tool for the conventional. - The roll pin of paper bagging machine was good with 5.3cm in terms of bagging precision. - There was no in grape quality between the paper bagging machine and the conventional method. - Disease infection and grape break was not in difference both treatments.

• Journal of Korean Society of Forest Science
• /
• v.98 no.6
• /
• pp.696-702
• /
• 2009

Yellow poplar (Liriodendron tulipifera L.) is an insect-pollinated tree species with large, perfect flower, and its seed sets average only about 10 percent naturally. In its controlled pollination, pollination bags are usually taken to prevent unwanted pollination, but bagging is an expensive and time-consuming process. Therefore, this study was conducted to determine the need of pollination bag by estimating how much unintended pollination would occur when different cross methods were applied. Five different pollination methods were applied as follows: 1) natural open pollination (i.e. insect pollination) as a reference, 2) self-pollination; no removing reproductive organs with bagging, 3) open pollination; emasculated(removing sepal, petal and stamen) without bagging, 4) controlled pollination; emasculated with bagging and 5) controlled pollination; emasculated without bagging. Very low value of full seed rate (0.2%) was observed in method 3, it was suggested that removing stamen and petal restrict the activity of pollen vectors like bee. Difference in the full seed rate between method 4 and method 5 was not significant (27.9% versus 24.0%, respectively). Consequently, controlled pollination without bagging might be an alternative method for extensive breeding and mass production of seeds in yellow poplar.

• International Journal of Computer Science & Network Security
• /
• v.24 no.1
• /
• pp.107-118
• /
• 2024

With the seamless growth of the technology, network usage requirements are expanding day by day. The majority of electronic devices are capable of communication, which strongly requires a secure and reliable network. Network-based intrusion detection systems (NIDS) is a new method for preventing and alerting computers and networks from attacks. Machine Learning is an emerging field that provides a variety of ways to implement effective network intrusion detection systems (NIDS). Bagging and Boosting are two ensemble ML techniques, renowned for better performance in the learning and classification process. In this paper, the study provides a detailed literature review of the past work done and proposed a novel ensemble approach to develop a NIDS system based on the voting method using bagging and boosting ensemble techniques. The test results demonstrate that the ensemble of bagging and boosting through voting exhibits the highest classification accuracy of 99.98% and a minimum false positive rate (FPR) on both datasets. Although the model building time is average which can be a tradeoff by processor speed.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.36 no.1
• /
• pp.9-18
• /
• 2023

Predicting the compressive strength of high-performance concrete (HPC) is challenging because of the use of additional cementitious materials; thus, the development of improved predictive models is essential. The purpose of this study was to develop an HPC compressive-strength prediction model using an ensemble machine-learning method of combined bagging and stacking techniques. The result is a new ensemble technique that integrates the existing ensemble methods of bagging and stacking to solve the problems of a single machine-learning model and improve the prediction performance of the model. The nonlinear regression, support vector machine, artificial neural network, and Gaussian process regression approaches were used as single machine-learning methods and bagging and stacking techniques as ensemble machine-learning methods. As a result, the model of the proposed method showed improved accuracy results compared with single machine-learning models, an individual bagging technique model, and a stacking technique model. This was confirmed through a comparison of four representative performance indicators, verifying the effectiveness of the method.

• The Korean Journal of Pesticide Science
• /
• v.21 no.1
• /
• pp.42-48
• /
• 2017

This study was conducted to compare the effect of bagging technique on the presence of pesticide residues in various parts of grape fruit (whole fruit, pulp and peel). The tested pesticides were diluted at 2,000 times and sprayed three times onto the crops at an interval of seven days and then they were collected at 0, 1, 3, 5 and 7 days after final application. Later, bagging/non-bagging samples were pre-treated with fruit, pulp and peel samples, respectively. Thiacloprid and lufenuron were not detected in any of the bagging samples. The thiacloprid residues of non-bagging samples in whole, peel and pulp samples were 0.47-1.09, 0.18-0.33 and 1.24-1.67 mg/kg, respectively. The lufenuron residues of non-bagging samples in whole fruit, peel and pulp samples were 0.16-0.62, < LOD-0.08 and 0.85-1.48 mg/kg, respectively. The biological half-lives of thiacloprid and lufenuron in whole fruit, peel and pulp of non-bagging samples were 5.7, 15.1 and 7.8 days and 4.0, 9.4 and 2.6 days, respectively. While the unbagged samples showed a sequential decrease in pesticide residues, this study concludes that bagging would be an effective method to protect the presence of thiacloprid and lufenuron residues in grape fruits.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.32 no.1
• /
• pp.33-41
• /
• 1996

In order to obtain the most favourable classification system for fishing gears, the problems in the existing systems were investigated and a new system in which the fishing method was adopted as the criterion of classification and the kinds of fishing gears were obtained by exchanging the word method into gear in the fishing methods classified newly for eliminating the problems was established. The new system to which the actual gears are arranged is as follows ; (1)Harvesting gear \circled1Plucking gears : Clamp, Tong, Wrench, etc. \circled2Sweeping gears : Push net, Coral sweep net, etc. \circled3Dredging gears : Hand dredge net, Boat dredge net, etc. (2)Sticking gears \circled1Shot sticking gears : Spear, Sharp plummet, Harpoon, etc. \circled2Pulled sticking gears : Gaff, Comb, Rake, Hook harrow, Jerking hook, etc. \circled3Left sticking gears : Rip - hook set line. (3)Angling gears \circled1Jerky angling gears (a)Single - jerky angling gears : Hand line, Pole line, etc. (b)Multiple - jerky angling gears : squid hook. \circled2Idly angling gears (a)Set angling gears : Set long line. (b)Drifted angling gears : Drift long line, Drift vertical line, etc. \circled3Dragged angling gears : Troll line. (4)Shelter gears : Eel tube, Webfoot - octopus pot, Octopus pot, etc. (5)Attracting gears : Fishing basket. (6)Cutoff gears : Wall, Screen net, Window net, etc. (7)Guiding gears \circled1Horizontally guiding gears : Triangular set net, Elliptic set net, Rectangular set net, Fish weir, etc. \circled2Vertically guiding gears : Pound net. \circled3Deeply guiding gears : Funnel net. (8)Receiving gears \circled1Jumping - fish receiving gears : Fish - receiving scoop net, Fish - receiving raft, etc. \circled2Drifting - fish receiving gears (a)Set drifting - fish receiving gears : Bamboo screen, Pillar stow net, Long stow net, etc. (b)Movable drifting - fish receiving gears : Stow net. (9)Bagging gears \circled1Drag - bagging gears (a)Bottom - drag bagging gears : Bottom otter trawl, Bottom beam trawl, Bottom pair trawl, etc. (b)Midwater - drag gagging gears : Midwater otter trawl, Midwater pair trawl, etc. (c)Surface - drag gagging gears : Anchovy drag net. \circled2Seine - bagging gears (a)Beach - seine bagging gears : Skimming scoop net, Beach seine, etc. (b)Boat - seine bagging gears : Boat seine, Danish seine, etc. \circled3Drive - bagging gears : Drive - in dustpan net, Inner drive - in net, etc. (10)Surrounding gears \circled1Incomplete surrounding gears : Lampara net, Ring net, etc. \circled2Complete surrounding gears : Purse seine, Round haul net, etc. (11)Covering gears \circled1Drop - type covering gears : Wooden cover, Lantern net, etc. \circled2Spread - type covering gears : Cast net. (12)Lifting gears \circled1Wait - lifting gears : Scoop net, Scrape net, etc. \circled2Gatherable lifting gears : Saury lift net, Anchovy lift net, etc. (13)Adherent gears \circled1Gilling gears (a)Set gilling gears : Bottom gill net, Floating gill net. (b)Drifted gilling gears : Drift gill net. (c)Encircled gilling gears : Encircled gill net. (d)Seine - gilling gears : Seining gill net. (e)Dragged gilling gears : Dragged gill net. \circled2Tangling gears (a)Set tangling gears : Double trammel net, Triple trammel net, etc. (b)Encircled tangling gears : Encircled tangle net. (c)Dragged tangling gears : Dragged tangle net. \circled3Restrainting gears (a)Drifted restrainting gears : Pocket net(Gen - type net). (b)Dragged restrainting gears : Dragged pocket net. (14)Sucking gears : Fish pumps.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.492-494
• /
• 2021

The decision tree technique is used as a classification technique in machine learning. However, the decision tree has a problem of consuming a lot of speed or resources due to the problem of overfitting. To solve this problem, there are bagging and boosting techniques. Bagging creates multiple samplings and models them using them, and boosting models the sampled data and adjusts weights to reduce overfitting. In addition, recently, techniques Xgboost have been introduced to improve performance. Therefore, in this paper, we collect wifi signal data for indoor positioning, apply it to the existing method and Xgboost, and perform performance evaluation through it.

• Journal of Intelligence and Information Systems
• /
• v.20 no.4
• /
• pp.121-139
• /
• 2014

Predicting corporate failure has been an important topic in accounting and finance. The costs associated with bankruptcy are high, so the accuracy of bankruptcy prediction is greatly important for financial institutions. Lots of researchers have dealt with the topic associated with bankruptcy prediction in the past three decades. The current research attempts to use ensemble models for improving the performance of bankruptcy prediction. Ensemble classification is to combine individually trained classifiers in order to gain more accurate prediction than individual models. Ensemble techniques are shown to be very useful for improving the generalization ability of the classifier. Bagging is the most commonly used methods for constructing ensemble classifiers. In bagging, the different training data subsets are randomly drawn with replacement from the original training dataset. Base classifiers are trained on the different bootstrap samples. Instance selection is to select critical instances while deleting and removing irrelevant and harmful instances from the original set. Instance selection and bagging are quite well known in data mining. However, few studies have dealt with the integration of instance selection and bagging. This study proposes an improved bagging ensemble based on instance selection using genetic algorithms (GA) for improving the performance of SVM. GA is an efficient optimization procedure based on the theory of natural selection and evolution. GA uses the idea of survival of the fittest by progressively accepting better solutions to the problems. GA searches by maintaining a population of solutions from which better solutions are created rather than making incremental changes to a single solution to the problem. The initial solution population is generated randomly and evolves into the next generation by genetic operators such as selection, crossover and mutation. The solutions coded by strings are evaluated by the fitness function. The proposed model consists of two phases: GA based Instance Selection and Instance based Bagging. In the first phase, GA is used to select optimal instance subset that is used as input data of bagging model. In this study, the chromosome is encoded as a form of binary string for the instance subset. In this phase, the population size was set to 100 while maximum number of generations was set to 150. We set the crossover rate and mutation rate to 0.7 and 0.1 respectively. We used the prediction accuracy of model as the fitness function of GA. SVM model is trained on training data set using the selected instance subset. The prediction accuracy of SVM model over test data set is used as fitness value in order to avoid overfitting. In the second phase, we used the optimal instance subset selected in the first phase as input data of bagging model. We used SVM model as base classifier for bagging ensemble. The majority voting scheme was used as a combining method in this study. This study applies the proposed model to the bankruptcy prediction problem using a real data set from Korean companies. The research data used in this study contains 1832 externally non-audited firms which filed for bankruptcy (916 cases) and non-bankruptcy (916 cases). Financial ratios categorized as stability, profitability, growth, activity and cash flow were investigated through literature review and basic statistical methods and we selected 8 financial ratios as the final input variables. We separated the whole data into three subsets as training, test and validation data set. In this study, we compared the proposed model with several comparative models including the simple individual SVM model, the simple bagging model and the instance selection based SVM model. The McNemar tests were used to examine whether the proposed model significantly outperforms the other models. The experimental results show that the proposed model outperforms the other models.