• Journal of Distribution Science
• /
• v.8 no.3
• /
• pp.17-26
• /
• 2010

Recently concerns have been raised due to the unbalanced supply of crops: the price of crops has been unstable and at one point the price went up so high that the word Agflation(agriculture+ inflation) was coined. Korea, in particular, is a small-sized country and needs to secure the stable supply of crops by investing in the produce importation at a national level. Investment in foreign produce importation is becoming more important as a measure for sufficient supply of crops, limited supply of domestic crops, weakened farming conditions worldwide, as well as recent changes in the use of crops due to the development of bio-fuels, influence of carbon emission on crops, the price increase in crops, and influx of foreign hot money. However, there are many problems with investing in foreign produce importation: lack of support from the government; lack of farming information and technology; difficulty in securing the capital; no immediate pay-off from the investment and insufficient management. Although foreign produce is originally more price-competitive than domestic produce, it loses its competiveness in the process of importation (due to high tariffs) and poor distribution system, which makes it difficult to sell in Korea. Therefore, investment in foreign produce importation is being questioned for feasibility; to make it possible, foreign produce must maintain the price-competitiveness. Especially, harvest of agricultural products depends on natural and geographical conditions of each country and those products have indigenous properties, so distribution system according to import and export of agricultural products should be treated more carefully than that of other industries. Distribution costs are differentiated into each item and include cost of sorting and wrapping, cost of wrapping materials, cost of domestic transport, cost of international transport and cost of clearing customs for import and export. So transporting and storing agricultural products generates considerable costs compared with other products. Also, due to upgrade of dietary life, needs for stability, taste and visible quality toward food including agricultural products are being raised and wrong way of storage causes decomposition of food and loss of freshness, making the storage more difficult than that in room temperature, so storage and transport in distribution of agricultural products needs specialty. In addition, because lack of specialty in distribution and circulation such as storage and wrapping does not solve limit factors in distance, the distribution and circulation has been limited to a form of import and export within short-distant region. Therefore, need for distribution out-sourcing which can satisfy specialty in managing distribution and circulation and it is needed to establish more effective distribution system. However, existing distribution system of agricultural products is exposed to various problems including problems in distribution channel, making distribution and strategy for distribution and those problems are as follows. First, in case of investment in overseas agricultural industry, stable supply of the products is difficult because areas of production are dispersed widely and influenced by outer factors due to including overseas distribution channels. Also, at the aspect of quality, standardization of products is difficult, distribution system is quite complicated and unreasonable due to long distribution channels according to international trade and financial and institutional support is not enough. Especially, there are quite a lot of ineffective factors including multi level distribution process, dramatic gap between production cost and customer's cost, lack of physical distribution facilities and difficulties in storage and transport due to lack of wrapping containers. Besides, because import and export of agricultural products has been manages under the company's own distribution according to transaction contract between manufacturers and exporting company, efficiency is low due to excessive investment in fixed costs and lack of specialty in dealing with agricultural products causes fall of value of products, showing the limit to lose price-competitiveness. Especially, because lack of specialty in distribution and circulation such as storage and wrapping does not solve limit factors in distance, the distribution and circulation has been limited to a form of import and export within short-distant region. Therefore, need for distribution out-sourcing which can satisfy specialty in managing distribution and circulation and it is needed to establish more effective distribution system. Second, among tangible and intangible services which promote the efficiency of the whole distribution, a function building distribution environment which includes distribution information, system for standard and inspection, distribution finance, system for diversification of risks, education and training, distribution administration and tax system is wanted. In general, such a function building distribution environment is difficult to be changed and supplement innovatively because its effect compared with investment does not appear immediately despite of its necessity. Especially, in case of distribution of agricultural products, as a function of collecting and distributing is performed individually through various channels, the importance of distribution information and standardization is getting more focus due to the problem of repetition of work and lack of specialty. Also, efficient management of distribution is quite difficult due to lack of professionals in distribution, so support to professional education is needed. Third, though effort to keep self-sufficiency ratio of staple food, rice is regarded as important at the government level, level of dependency on overseas of others crops is high. Therefore, plan for stable securing food resources aside from staple food is also necessary. Especially, governmental organizations of agricultural products distribution in Korea are production-centered and have unreasonable structure whose function at the aspect of distribution and consumption is quite insufficient. And development of new distribution channels which can deal with changes in distribution environment and they do not achieve actual results of strategy for distribution due to non-positive strategy for price distribution. That is, it implies the possibility that base for supply will become vulnerable because it does not mediate appropriate interests on total distribution channels such as manufacturers, wholesale dealers and vendors by emphasizing consumer protection excessively in the distribution of agricultural products. Therefore, this study examined fundamental concept and actual situation for our investment to overseas agriculture, drew necessities, considerations, problems, etc. of overseas agricultural investment and suggested improvements at the level of distribution for price competitiveness of agricultural products cultivated in overseas under five aspects; government's indirect support, distribution's modernization and distribution information function's strengthening, government's political support for distribution facility, transportation route, load and unloading works' improvement, price competitiveness' securing, professional manpower's cultivation by education and training, etc. Here are some suggestions for foreign produce importation. First, the government should conduct a survey on the current distribution channels and analyze the situation to establish a measure for long-term development plans. By providing each agricultural area with a guideline for planning appropriate production of crops, the government can help farmers be ready for importation, and prevent them from producing same crops all at the same time. Government can sign an MOU with the foreign government and promote the importation so that the development of agricultural resources can be stable and steady. Second, the government can establish a strategy for an effective distribution system by providing farmers and agriculture-related workers with the distribution information such as price, production, demand, market structure and location, feature of each crop, and etc. In order for such distribution system to become feasible, the government needs to reconstruct the current distribution system, designate a public organization for providing distribution information and set the criteria for level of produce quality, trade units, and package units. Third, the government should provide financial support and a policy to seek an efficient distribution channel for foreign produce to be delivered fresh: the government should expand distribution facilities (for selecting, packaging, storing, and processing) and transportation vehicles while modernizing old facilities. There should be another policy to improve the efficiency of unloading, and to lower the cost of distribution. Fourth, it is necessary to enact a new law covering exceptional cases for importing produce in order to maintain the price competitiveness; currently the high tariffs is keeping the imported produce from being distributed domestically. However, the new adjustment should be made carefully within the WTO regulations since it can create a problem from giving preferential tariffs. The government can also simplify the distribution channels in order to reduce the cost in the distribution process. Fifth, the government should educate distributors to raise the efficiency and to modernize the distribution system. It is necessary to develop human resources by educating people regarding the foreign agricultural environment, the produce quality, management skills, and by introducing some successful cases in advanced countries.

• Journal of Intelligence and Information Systems
• /
• v.23 no.2
• /
• pp.71-88
• /
• 2017

Language models were originally developed for speech recognition and language processing. Using a set of example sentences, a language model predicts the next word or character based on sequential input data. N-gram models have been widely used but this model cannot model the correlation between the input units efficiently since it is a probabilistic model which are based on the frequency of each unit in the training set. Recently, as the deep learning algorithm has been developed, a recurrent neural network (RNN) model and a long short-term memory (LSTM) model have been widely used for the neural language model (Ahn, 2016; Kim et al., 2016; Lee et al., 2016). These models can reflect dependency between the objects that are entered sequentially into the model (Gers and Schmidhuber, 2001; Mikolov et al., 2010; Sundermeyer et al., 2012). In order to learning the neural language model, texts need to be decomposed into words or morphemes. Since, however, a training set of sentences includes a huge number of words or morphemes in general, the size of dictionary is very large and so it increases model complexity. In addition, word-level or morpheme-level models are able to generate vocabularies only which are contained in the training set. Furthermore, with highly morphological languages such as Turkish, Hungarian, Russian, Finnish or Korean, morpheme analyzers have more chance to cause errors in decomposition process (Lankinen et al., 2016). Therefore, this paper proposes a phoneme-level language model for Korean language based on LSTM models. A phoneme such as a vowel or a consonant is the smallest unit that comprises Korean texts. We construct the language model using three or four LSTM layers. Each model was trained using Stochastic Gradient Algorithm and more advanced optimization algorithms such as Adagrad, RMSprop, Adadelta, Adam, Adamax, and Nadam. Simulation study was done with Old Testament texts using a deep learning package Keras based the Theano. After pre-processing the texts, the dataset included 74 of unique characters including vowels, consonants, and punctuation marks. Then we constructed an input vector with 20 consecutive characters and an output with a following 21st character. Finally, total 1,023,411 sets of input-output vectors were included in the dataset and we divided them into training, validation, testsets with proportion 70:15:15. All the simulation were conducted on a system equipped with an Intel Xeon CPU (16 cores) and a NVIDIA GeForce GTX 1080 GPU. We compared the loss function evaluated for the validation set, the perplexity evaluated for the test set, and the time to be taken for training each model. As a result, all the optimization algorithms but the stochastic gradient algorithm showed similar validation loss and perplexity, which are clearly superior to those of the stochastic gradient algorithm. The stochastic gradient algorithm took the longest time to be trained for both 3- and 4-LSTM models. On average, the 4-LSTM layer model took 69% longer training time than the 3-LSTM layer model. However, the validation loss and perplexity were not improved significantly or became even worse for specific conditions. On the other hand, when comparing the automatically generated sentences, the 4-LSTM layer model tended to generate the sentences which are closer to the natural language than the 3-LSTM model. Although there were slight differences in the completeness of the generated sentences between the models, the sentence generation performance was quite satisfactory in any simulation conditions: they generated only legitimate Korean letters and the use of postposition and the conjugation of verbs were almost perfect in the sense of grammar. The results of this study are expected to be widely used for the processing of Korean language in the field of language processing and speech recognition, which are the basis of artificial intelligence systems.

• Journal of Intelligence and Information Systems
• /
• v.18 no.2
• /
• pp.29-45
• /
• 2012

Bond rating is regarded as an important event for measuring financial risk of companies and for determining the investment returns of investors. As a result, it has been a popular research topic for researchers to predict companies' credit ratings by applying statistical and machine learning techniques. The statistical techniques, including multiple regression, multiple discriminant analysis (MDA), logistic models (LOGIT), and probit analysis, have been traditionally used in bond rating. However, one major drawback is that it should be based on strict assumptions. Such strict assumptions include linearity, normality, independence among predictor variables and pre-existing functional forms relating the criterion variablesand the predictor variables. Those strict assumptions of traditional statistics have limited their application to the real world. Machine learning techniques also used in bond rating prediction models include decision trees (DT), neural networks (NN), and Support Vector Machine (SVM). Especially, SVM is recognized as a new and promising classification and regression analysis method. SVM learns a separating hyperplane that can maximize the margin between two categories. SVM is simple enough to be analyzed mathematical, and leads to high performance in practical applications. SVM implements the structuralrisk minimization principle and searches to minimize an upper bound of the generalization error. In addition, the solution of SVM may be a global optimum and thus, overfitting is unlikely to occur with SVM. In addition, SVM does not require too many data sample for training since it builds prediction models by only using some representative sample near the boundaries called support vectors. A number of experimental researches have indicated that SVM has been successfully applied in a variety of pattern recognition fields. However, there are three major drawbacks that can be potential causes for degrading SVM's performance. First, SVM is originally proposed for solving binary-class classification problems. Methods for combining SVMs for multi-class classification such as One-Against-One, One-Against-All have been proposed, but they do not improve the performance in multi-class classification problem as much as SVM for binary-class classification. Second, approximation algorithms (e.g. decomposition methods, sequential minimal optimization algorithm) could be used for effective multi-class computation to reduce computation time, but it could deteriorate classification performance. Third, the difficulty in multi-class prediction problems is in data imbalance problem that can occur when the number of instances in one class greatly outnumbers the number of instances in the other class. Such data sets often cause a default classifier to be built due to skewed boundary and thus the reduction in the classification accuracy of such a classifier. SVM ensemble learning is one of machine learning methods to cope with the above drawbacks. Ensemble learning is a method for improving the performance of classification and prediction algorithms. AdaBoost is one of the widely used ensemble learning techniques. It constructs a composite classifier by sequentially training classifiers while increasing weight on the misclassified observations through iterations. The observations that are incorrectly predicted by previous classifiers are chosen more often than examples that are correctly predicted. Thus Boosting attempts to produce new classifiers that are better able to predict examples for which the current ensemble's performance is poor. In this way, it can reinforce the training of the misclassified observations of the minority class. This paper proposes a multiclass Geometric Mean-based Boosting (MGM-Boost) to resolve multiclass prediction problem. Since MGM-Boost introduces the notion of geometric mean into AdaBoost, it can perform learning process considering the geometric mean-based accuracy and errors of multiclass. This study applies MGM-Boost to the real-world bond rating case for Korean companies to examine the feasibility of MGM-Boost. 10-fold cross validations for threetimes with different random seeds are performed in order to ensure that the comparison among three different classifiers does not happen by chance. For each of 10-fold cross validation, the entire data set is first partitioned into tenequal-sized sets, and then each set is in turn used as the test set while the classifier trains on the other nine sets. That is, cross-validated folds have been tested independently of each algorithm. Through these steps, we have obtained the results for classifiers on each of the 30 experiments. In the comparison of arithmetic mean-based prediction accuracy between individual classifiers, MGM-Boost (52.95%) shows higher prediction accuracy than both AdaBoost (51.69%) and SVM (49.47%). MGM-Boost (28.12%) also shows the higher prediction accuracy than AdaBoost (24.65%) and SVM (15.42%)in terms of geometric mean-based prediction accuracy. T-test is used to examine whether the performance of each classifiers for 30 folds is significantly different. The results indicate that performance of MGM-Boost is significantly different from AdaBoost and SVM classifiers at 1% level. These results mean that MGM-Boost can provide robust and stable solutions to multi-classproblems such as bond rating.