• Journal of Intelligence and Information Systems
• /
• 제27권1호
• /
• pp.83-102
• /
• 2021

The government recently announced various policies for developing big-data and artificial intelligence fields to provide a great opportunity to the public with respect to disclosure of high-quality data within public institutions. KSURE(Korea Trade Insurance Corporation) is a major public institution for financial policy in Korea, and thus the company is strongly committed to backing export companies with various systems. Nevertheless, there are still fewer cases of realized business model based on big-data analyses. In this situation, this paper aims to develop a new business model which can be applied to an ex-ante prediction for the likelihood of the insurance accident of credit guarantee. We utilize internal data from KSURE which supports export companies in Korea and apply machine learning models. Then, we conduct performance comparison among the predictive models including Logistic Regression, Random Forest, XGBoost, LightGBM, and DNN(Deep Neural Network). For decades, many researchers have tried to find better models which can help to predict bankruptcy since the ex-ante prediction is crucial for corporate managers, investors, creditors, and other stakeholders. The development of the prediction for financial distress or bankruptcy was originated from Smith(1930), Fitzpatrick(1932), or Merwin(1942). One of the most famous models is the Altman's Z-score model(Altman, 1968) which was based on the multiple discriminant analysis. This model is widely used in both research and practice by this time. The author suggests the score model that utilizes five key financial ratios to predict the probability of bankruptcy in the next two years. Ohlson(1980) introduces logit model to complement some limitations of previous models. Furthermore, Elmer and Borowski(1988) develop and examine a rule-based, automated system which conducts the financial analysis of savings and loans. Since the 1980s, researchers in Korea have started to examine analyses on the prediction of financial distress or bankruptcy. Kim(1987) analyzes financial ratios and develops the prediction model. Also, Han et al.(1995, 1996, 1997, 2003, 2005, 2006) construct the prediction model using various techniques including artificial neural network. Yang(1996) introduces multiple discriminant analysis and logit model. Besides, Kim and Kim(2001) utilize artificial neural network techniques for ex-ante prediction of insolvent enterprises. After that, many scholars have been trying to predict financial distress or bankruptcy more precisely based on diverse models such as Random Forest or SVM. One major distinction of our research from the previous research is that we focus on examining the predicted probability of default for each sample case, not only on investigating the classification accuracy of each model for the entire sample. Most predictive models in this paper show that the level of the accuracy of classification is about 70% based on the entire sample. To be specific, LightGBM model shows the highest accuracy of 71.1% and Logit model indicates the lowest accuracy of 69%. However, we confirm that there are open to multiple interpretations. In the context of the business, we have to put more emphasis on efforts to minimize type 2 error which causes more harmful operating losses for the guaranty company. Thus, we also compare the classification accuracy by splitting predicted probability of the default into ten equal intervals. When we examine the classification accuracy for each interval, Logit model has the highest accuracy of 100% for 0~10% of the predicted probability of the default, however, Logit model has a relatively lower accuracy of 61.5% for 90~100% of the predicted probability of the default. On the other hand, Random Forest, XGBoost, LightGBM, and DNN indicate more desirable results since they indicate a higher level of accuracy for both 0~10% and 90~100% of the predicted probability of the default but have a lower level of accuracy around 50% of the predicted probability of the default. When it comes to the distribution of samples for each predicted probability of the default, both LightGBM and XGBoost models have a relatively large number of samples for both 0~10% and 90~100% of the predicted probability of the default. Although Random Forest model has an advantage with regard to the perspective of classification accuracy with small number of cases, LightGBM or XGBoost could become a more desirable model since they classify large number of cases into the two extreme intervals of the predicted probability of the default, even allowing for their relatively low classification accuracy. Considering the importance of type 2 error and total prediction accuracy, XGBoost and DNN show superior performance. Next, Random Forest and LightGBM show good results, but logistic regression shows the worst performance. However, each predictive model has a comparative advantage in terms of various evaluation standards. For instance, Random Forest model shows almost 100% accuracy for samples which are expected to have a high level of the probability of default. Collectively, we can construct more comprehensive ensemble models which contain multiple classification machine learning models and conduct majority voting for maximizing its overall performance.

• Journal of Intelligence and Information Systems
• /
• 제27권1호
• /
• pp.103-128
• /
• 2021

Recently, investors' interest and the influence of stock-related information dissemination are being considered as significant factors that explain stock returns and volume. Besides, companies that develop, distribute, or utilize innovative new technologies such as artificial intelligence have a problem that it is difficult to accurately predict a company's future stock returns and volatility due to macro-environment and market uncertainty. Market uncertainty is recognized as an obstacle to the activation and spread of artificial intelligence technology, so research is needed to mitigate this. Hence, the purpose of this study is to propose a machine learning model that predicts the volatility of a company's stock price by using the internet search volume of artificial intelligence-related technology keywords as a measure of the interest of investors. To this end, for predicting the stock market, we using the VAR(Vector Auto Regression) and deep neural network LSTM (Long Short-Term Memory). And the stock price prediction performance using keyword search volume is compared according to the technology's social acceptance stage. In addition, we also conduct the analysis of sub-technology of artificial intelligence technology to examine the change in the search volume of detailed technology keywords according to the technology acceptance stage and the effect of interest in specific technology on the stock market forecast. To this end, in this study, the words artificial intelligence, deep learning, machine learning were selected as keywords. Next, we investigated how many keywords each week appeared in online documents for five years from January 1, 2015, to December 31, 2019. The stock price and transaction volume data of KOSDAQ listed companies were also collected and used for analysis. As a result, we found that the keyword search volume for artificial intelligence technology increased as the social acceptance of artificial intelligence technology increased. In particular, starting from AlphaGo Shock, the keyword search volume for artificial intelligence itself and detailed technologies such as machine learning and deep learning appeared to increase. Also, the keyword search volume for artificial intelligence technology increases as the social acceptance stage progresses. It showed high accuracy, and it was confirmed that the acceptance stages showing the best prediction performance were different for each keyword. As a result of stock price prediction based on keyword search volume for each social acceptance stage of artificial intelligence technologies classified in this study, the awareness stage's prediction accuracy was found to be the highest. The prediction accuracy was different according to the keywords used in the stock price prediction model for each social acceptance stage. Therefore, when constructing a stock price prediction model using technology keywords, it is necessary to consider social acceptance of the technology and sub-technology classification. The results of this study provide the following implications. First, to predict the return on investment for companies based on innovative technology, it is most important to capture the recognition stage in which public interest rapidly increases in social acceptance of the technology. Second, the change in keyword search volume and the accuracy of the prediction model varies according to the social acceptance of technology should be considered in developing a Decision Support System for investment such as the big data-based Robo-advisor recently introduced by the financial sector.

• Journal of KIISE
• /
• 제44권5호
• /
• pp.503-509
• /
• 2017

In this paper, the copy mechanism and input feeding are applied to recurrent neural network(RNN)-search model in a Korean-document summarization in an end-to-end manner. In addition, the performances of the document summarizations are compared according to the model and the tokenization format; accordingly, the syllable-unit, morpheme-unit, and hybrid-unit tokenization formats are compared. For the experiments, Internet newspaper articles were collected to construct a Korean-document summary data set (train set: 30291 documents; development set: 3786 documents; test set: 3705 documents). When the format was tokenized as the morpheme-unit, the models with the input feeding and the copy mechanism showed the highest performances of ROUGE-1 35.92, ROUGE-2 15.37, and ROUGE-L 29.45.

• Journal of KIISE
• /
• 제44권1호
• /
• pp.36-43
• /
• 2017

Syntactic information represents the dependency relation between predicates and arguments, and it is helpful for improving the performance of Semantic Role Labeling systems. However, syntax analysis can cause computational overhead and inherit incorrect syntactic information. To solve this problem, we exclude syntactic information and use only morpheme information to construct Semantic Role Labeling systems. In this study, we propose an end-to-end SRL system that only uses morpheme information with Stacked Bidirectional LSTM-CRFs model by extending the LSTM RNN that is suitable for sequence labeling problem. Our experimental results show that our proposed model has better performance, as compare to other models.

• Journal of Broadcast Engineering
• /
• 제23권5호
• /
• pp.669-681
• /
• 2018

CNN (Convolution Neural Network) is one of the most important techniques to identify the kind of objects in the captured pictures. Whereas the conventional models have been used for low resolution images, the technique to recognize the high resolution images becomes crucial in the field of artificial intelligence. In this paper, we proposed an efficient CNN model based on dilated convolution and thresholding techniques to increase the recognition ratio and to decrease the computational complexity. The simulation results show that the proposed algorithm outperforms the conventional method and the thresholding technique enhances the performance of the proposed model.

• KIISE Transactions on Computing Practices
• /
• 제23권9호
• /
• pp.542-549
• /
• 2017

Sequence-to-sequence models and similar pointer networks suffer from performance degradation when an input is composed of multiple sentences or when the length of the input sentence is long. To solve this problem, this paper proposes a hierarchical pointer network model that uses both the word level and sentence level information to encode input sequences composed of several sentences at the word level and sentence level. We propose a hierarchical pointer network based coreference resolution that performs a coreference resolution for all mentions. The experimental results show that the proposed model has a precision of 87.07%, recall of 65.39% and CoNLL F1 74.61%, which is an improvement of 21.83% compared to an existing rule-based model.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권7호
• /
• pp.3268-3283
• /
• 2018

With the advent of the Information Age, the source identification of digital images, as a part of digital image forensics, has attracted increasing attention. Therefore, an effective technique to identify the source of digital images is urgently needed at this stage. In this paper, first, we study and implement some previous work on image source identification based on sensor pattern noise, such as the Lukas method, principal component analysis method and the random subspace method. Second, to extract a purer sensor pattern noise, we propose a sample selection method to improve the random subspace method. By analyzing the image texture feature, we select a patch with less complexity to extract more reliable sensor pattern noise, which improves the accuracy of identification. Finally, experiment results reveal that the proposed sample selection method can extract a purer sensor pattern noise, which further improves the accuracy of image source identification. At the same time, this approach is less complicated than the deep learning models and is close to the most advanced performance.

• Communications for Statistical Applications and Methods
• /
• 제27권3호
• /
• pp.313-326
• /
• 2020

Deep Learning is the most important key to the development of Artificial Intelligence (AI). There are several distinguishable architectures of neural networks such as MLP, CNN, and RNN. Among them, we try to understand one of the main architectures called Recurrent Neural Network (RNN) that differs from other networks in handling sequential data, including time series and texts. As one of the main tasks recently in Natural Language Processing (NLP), we consider Neural Machine Translation (NMT) using RNNs. We also summarize fundamental structures of the recurrent networks, and some topics of representing natural words to reasonable numeric vectors. We organize topics to understand estimation procedures from representing input source sequences to predict target translated sequences. In addition, we apply multiple translation models with Gated Recurrent Unites (GRUs) in Keras on English-Korean sentences that contain about 26,000 pairwise sequences in total from two different corpora, colloquialism and news. We verified some crucial factors that influence the quality of training. We found that loss decreases with more recurrent dimensions and using bidirectional RNN in the encoder when dealing with short sequences. We also computed BLEU scores which are the main measures of the translation performance, and compared them with the score from Google Translate using the same test sentences. We sum up some difficulties when training a proper translation model as well as dealing with Korean language. The use of Keras in Python for overall tasks from processing raw texts to evaluating the translation model also allows us to include some useful functions and vocabulary libraries as well.

• Journal of Information Processing Systems
• /
• 제16권1호
• /
• pp.224-237
• /
• 2020

Recognition of banknote serial number is one of the important functions for intelligent banknote counter implementation and can be used for various purposes. However, the previous character recognition method is limited to use due to the font type of the banknote serial number, the variation problem by the solid status, and the recognition speed issue. In this paper, we propose an aspect ratio based character region segmentation and a convolutional neural network (CNN) based banknote serial number recognition method. In order to detect the character region, the character area is determined based on the aspect ratio of each character in the serial number candidate area after the banknote area detection and de-skewing process is performed. Then, we designed and compared four types of CNN models and determined the best model for serial number recognition. Experimental results showed that the recognition accuracy of each character was 99.85%. In addition, it was confirmed that the recognition performance is improved as a result of performing data augmentation. The banknote used in the experiment is Indian rupee, which is badly soiled and the font of characters is unusual, therefore it can be regarded to have good performance. Recognition speed was also enough to run in real time on a device that counts 800 banknotes per minute.

• Annual Conference of KIPS
• /
• 한국정보처리학회 2019년도 춘계학술발표대회
• /
• pp.419-422
• /
• 2019

The recent increase in single-person households and family income has led to an increase in the number of pet owners. However, due to the owners' difficulty to communicate with them for 24 hours, pets, and especially dogs, tend to display unwanted behavior that can be harmful to themselves and their environment when left alone. Therefore, detecting those behaviors when the owner is absent is necessary to suppress them and prevent any damage. In this paper, we propose a camera-based system that detects a set of normal and unwanted behaviors using deep learning algorithms to monitor dogs when left alone at home. The frames collected from the camera are arranged into sequences of RGB frames and their corresponding optical flow sequences, and then features are extracted from each data flow using pre-trained VGG-16 models. The extracted features from each sequence are concatenated and input to a bi-directional LSTM network that classifies the dog action into one of the targeted classes. The experimental results show that our method achieves a good performance exceeding 0.9 in precision, recall and f-1 score.