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

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

• KDI Journal of Economic Policy
• /
• v.12 no.3
• /
• pp.71-94
• /
• 1990

This paper reviews the rationale of monetary and banking intervention by an outside authority, either the government or the central bank, and seeks to delineate clearly the optimal limits to the monetary and banking deregulation currently underway in Korea as well as on a global scale. Furthermore, this paper seeks to establish an objective and balanced view on the role of the central bank, especially in light of the current discussion on the restructuring of Korea's central bank, which has been severely contaminated by interest-group politics. The discussion begins with the recognition that the modern free banking school and the new monetary economics are becoming formidable challenges to the traditional role of the government or the central bank in the monetary and banking sector. The paper reviews six arguments that have traditionally been presented to support intervention: (1) the possibility of an over-issue of bank notes under free banking instead of central banking; (2) externalities in and the public good nature of the use of money; (3) economies of scale and natural monopoly in producing money; (4) the need for macro stabilization policy due to the instability of the real sector; (5) the external effects of bank failure due to the inherent instability of the existing banking system; and (6) protection for small banknote users and depositors. Based on an analysis of the above arguments, the paper speculates on the optimal role of the government or central bank in the monetary and banking system and the optimal degree of monetary and banking deregulation. By contrast to the arguments for free banking or laissez-faire monetary systems, which become fashionable in recent years, monopoly and intervention by the government or central bank in the outside money system can be both necessary and optimal. In this case, of course, an over-issue of fiat money may be possible due to political considerations, but this issue is beyond the scope of this paper. On the other hand, the issue of inside monies based on outside money could indeed be provided for optimally under market competition by private institutions. A competitive system in issuing inside monies would help realize, to the maxim urn extent possible, external economies generated by using a single outside money. According to this reasoning, free banking activities will prevail in the inside money system, while a government monopoly will prevail in the outside money system. This speculation, then, also implies that the monetary and banking deregulation currently underway should and most likely will be limited to the inside money system, which could be liberalized to the fullest degree. It is also implied that it will be impractical to deregulate the outside money system and to allow market competition to provide outside money, in accordance with the arguments of the free banking school and the new monetary economics. Furthermore, the role of the government or central bank in this new environment will not be significantly different from their current roles. As far as the supply of fiat money continues to be monopolized by the government, the control of the supply of base money and such related responsibilities as monetary policy (argument(4)) and the lender of the last resort (argument (5)) will naturally be assigned to the outside money supplier. However, a mechanism for controlling an over-issue of fiat money by a monopolistic supplier will definitely be called for (argument(1)). A monetary policy based on a certain policy rule could be one possibility. More importantly, the deregulation of the inside money system would further increase the systemic risk inherent in the current fractional banking system, while enhancing the efficiency of the system (argument (5)). In this context, the role of the lender of the last resort would again become an instrument of paramount importance in alleviating liquidity crises in the early stages, thereby disallowing the possibility of a widespread bank run. Similarly, prudential banking supervision would also help maintain the safety and soundness of the fully deregulated banking system. These functions would also help protect depositors from losses due to bank failures (argument (6)). Finally, these speculations suggest that government or central bank authorities have probably been too conservative on the issue of the deregulation of the financial system, beyond the caution necessary to preserve system safety. Rather, only the fullest deregulation of the inside money system seems to guarantee the maximum enjoyment of external economies in the single outside money system.