• Journal of Intelligence and Information Systems
• /
• v.27 no.1
• /
• 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 Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.139-161
• /
• 2019

The utilization of the e-commerce market has become a common life style in today. It has become important part to know where and how to make reasonable purchases of good quality products for customers. This change in purchase psychology tends to make it difficult for customers to make purchasing decisions in vast amounts of information. In this case, the recommendation system has the effect of reducing the cost of information retrieval and improving the satisfaction by analyzing the purchasing behavior of the customer. Amazon and Netflix are considered to be the well-known examples of sales marketing using the recommendation system. In the case of Amazon, 60% of the recommendation is made by purchasing goods, and 35% of the sales increase was achieved. Netflix, on the other hand, found that 75% of movie recommendations were made using services. This personalization technique is considered to be one of the key strategies for one-to-one marketing that can be useful in online markets where salespeople do not exist. Recommendation techniques that are mainly used in recommendation systems today include collaborative filtering and content-based filtering. Furthermore, hybrid techniques and association rules that use these techniques in combination are also being used in various fields. Of these, collaborative filtering recommendation techniques are the most popular today. Collaborative filtering is a method of recommending products preferred by neighbors who have similar preferences or purchasing behavior, based on the assumption that users who have exhibited similar tendencies in purchasing or evaluating products in the past will have a similar tendency to other products. However, most of the existed systems are recommended only within the same category of products such as books and movies. This is because the recommendation system estimates the purchase satisfaction about new item which have never been bought yet using customer's purchase rating points of a similar commodity based on the transaction data. In addition, there is a problem about the reliability of purchase ratings used in the recommendation system. Reliability of customer purchase ratings is causing serious problems. In particular, 'Compensatory Review' refers to the intentional manipulation of a customer purchase rating by a company intervention. In fact, Amazon has been hard-pressed for these "compassionate reviews" since 2016 and has worked hard to reduce false information and increase credibility. The survey showed that the average rating for products with 'Compensated Review' was higher than those without 'Compensation Review'. And it turns out that 'Compensatory Review' is about 12 times less likely to give the lowest rating, and about 4 times less likely to leave a critical opinion. As such, customer purchase ratings are full of various noises. This problem is directly related to the performance of recommendation systems aimed at maximizing profits by attracting highly satisfied customers in most e-commerce transactions. In this study, we propose the possibility of using new indicators that can objectively substitute existing customer 's purchase ratings by using RFM multi-dimensional analysis technique to solve a series of problems. RFM multi-dimensional analysis technique is the most widely used analytical method in customer relationship management marketing(CRM), and is a data analysis method for selecting customers who are likely to purchase goods. As a result of verifying the actual purchase history data using the relevant index, the accuracy was as high as about 55%. This is a result of recommending a total of 4,386 different types of products that have never been bought before, thus the verification result means relatively high accuracy and utilization value. And this study suggests the possibility of general recommendation system that can be applied to various offline product data. If additional data is acquired in the future, the accuracy of the proposed recommendation system can be improved.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.15 no.4
• /
• pp.26-41
• /
• 2012

It is necessary to manage, forecast and prepare agricultural production based on accurate and up-to-date information in order to cope with the climate change and its impacts such as global warming, floods and droughts. This study examined the applicability as well as challenges of the object-based image analysis method for developing a land cover image classification algorithm, which can support the fast thematic mapping of wide agricultural areas on a regional scale. In order to test the applicability of RapidEye's multi-temporal spectral information for differentiating agricultural land cover types, the integration of other GIS data was minimized. Under this circumstance, the land cover classification accuracy at the study area of Kimje ($1300km^2$) was 80.3%. The geometric resolution of RapidEye, 6.5m showed the possibility to derive the spatial features of agricultural land use generally cultivated on a small scale in Korea. The object-based image analysis method can realize the expert knowledge in various ways during the classification process, so that the application of spectral image information can be optimized. An additional advantage is that the already developed classification algorithm can be stored, edited with variables in detail with regard to analytical purpose, and may be applied to other images as well as other regions. However, the segmentation process, which is fundamental for the object-based image classification, often cannot be explained quantitatively. Therefore, it is necessary to draw the best results based on expert's empirical and scientific knowledge.

• Journal of the East Asian Society of Dietary Life
• /
• v.17 no.4
• /
• pp.571-579
• /
• 2007

The aim of this study was to predict the development process of the Korean food service industry by forecasting the per capita GDP. Forecasting the GDP, involved two primary approaches. One was related to looking at the Korean food service industry's situation by per capita GDP and comparing it to that of the US and Japan. The other was to predict food service industry projections in Korea by quantitative forecasting models. Holt's simple exponential smoothing method and new types of the series models(Damped trend exponential smoothing method), were employed to predict the per capita GDP. The accuracy of the models was measured by MAPE. The empirical results of the forecasting models indicate that the three time series models performed fairly well. Of these Damped trend Damped trend exponential smoothing performed best with the lowest MAPE(9.9%). The results show that the time for reaching a per capita GDP level of $20,000 was 2008 with the Damped trend model and 2009 with the Holt model. Moreover, we found that a per capita GDP level of $30,000 will be achieved in 2012 from the Damped trend model and in 2013 from the Holt model. Within this study, the implications for the Korean food service industry are further discussed. It was predicted there will be a stabilization period in 2008 or 2009 in Korea with achievement of a per capita GDP of $20,000. At this time, major food service industry companies will need to invest in equipment toy external growth and there will be industry trends toward ethnic food and theme restaurants. Also, if a per capita GDP of $30,000 is achieved by 2012 or 2013, the Korean food industry will need to be highly responsive. Therefore, food industry companies should forecast and study customer values and prepare for changes.

• Journal of the Korean earth science society
• /
• v.35 no.4
• /
• pp.215-224
• /
• 2014

Data assimilation effect of mobile rawinsonde observation was evaluated using Unified Model (UM) with a Three-Dimensional Variational (3DVAR) data assimilation system during the intensive observation program of 2013 summer season (rainy season: 20 June-7 July 2013, heavy rain period: 8 July-30 July 2013). The analysis was performed by two sets of simulation experiments: (1) ConTroL experiment (CTL) with observation data provided by Korea Meteorological Administration (KMA) and (2) Observing System Experiment (OSE) including both KMA and mobile rawinsonde observation data. In the model verification during the rainy season, there were no distinctive differences for 500 hPa geopotential height, 850 hPa air temperature, and 300 hPa wind speed between CTL and OSE simulation due to data limitation (0000 and 1200 UTC only) at stationary rawinsonde stations. In contrast, precipitation verification using the hourly accumulated precipitation data of Automatic Synoptic Observation System (ASOS) showed that Equivalent Threat Score (ETS) of the OSE was improved by about 2% compared with that of the CTL. For cases having a positive effect of the OSE simulation, ETS of the OSE showed a significantly higher improvement (up to 41%) than that of the CTL. This estimation thus suggests that the use of mobile rawinsonde observation data using UM 3DVAR could be reasonable enough to assess the improvement of prediction accuracy.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.9 no.3
• /
• pp.209-215
• /
• 2007

To improve the accuracy of frost warning system for pear orchard in a complex terrain in Naju area, the daily minimum temperature forecasted by Korea Meteorological Administration (KMA) was interpolated using a regional climate model based on topoclimatic estimation and optimum scale interpolation from 2004 to 2005. Based on the validation experiments done for three pear orchards in the spring of 2004, the results showed a good agreement between the observed and predicted values, resulting in improved predictability compared to the forecast from Korea Meteorological Administration. The differences between the observed and the predicted temperatures were $-2.1{\sim}2.7^{\circ}C$ (on average $-0.4^{\circ}C$) in the valley, $-1.6{\sim}2.7^{\circ}C$ (on average $-0.4^{\circ}C$) in the riverside and $-1.1{\sim}3.5^{\circ}C$ (on average $0.6^{\circ}C$) in the hills. Notably, the errors have been reduced significantly for the valley and riverside areas that are more affected by the cold air drainage and more susceptible to frost damage than hills.

• 한국방재학회:학술대회논문집
• /
• 2011.02a
• /
• pp.172-172
• /
• 2011

The interaction between land surface and atmosphere is essentially affected by hydrometeorological variables including soil moisture. Accurate estimation of soil moisture at spatial and temporal scales is crucial to better understand its roles to the weather systems. The KLDAS(Korea Land Data Assimilation System) is a regional, specifically Korea peninsula land surface information systems. As other prior land data assimilation systems, this can provide initial soil field information which can be used in atmospheric simulations. For this study, as an enabling high-resolution tool, weather research and forecasting(WRF-ARW) model is applied to produce precipitation data using GFS(Global Forecast System) with GFS embedded and KLDAS soil moisture information as initialization data. WRF-ARW generates precipitation data for a specific region using different parameters in physics options. The produced precipitation data will be employed for simulations of Hydrological Models such as HEC(Hydrologic Engineering Center) - HMS(Hydrologic Modeling System) as predefined input data for selected regional water responses. The purpose of this study is to show the impact of a hydrometeorological variable such as soil moisture in KLDAS on hydrological consequences in Korea peninsula. The study region, Chongmi River Basin, is located in the center of Korea Peninsular. This has 60.8Km river length and 17.01% slope. This region mostly consists of farming field however the chosen study area placed in mountainous area. The length of river basin perimeter is 185Km and the average width of river is 9.53 meter with 676 meter highest elevation in this region. We have four different observation locations : Sulsung, Taepyung, Samjook, and Sangkeug observatoriesn, This watershed is selected as a tentative research location and continuously studied for getting hydrological effects from land surface information. Simulations for a real regional storm case(June 17~ June 25, 2006) are executed. WRF-ARW for this case study used WSM6 as a micro physics, Kain-Fritcsch Scheme for cumulus scheme, and YSU scheme for planetary boundary layer. The results of WRF simulations generate excellent precipitation data in terms of peak precipitation and date, and the pattern of daily precipitation for four locations. For Sankeug observatory, WRF overestimated precipitation approximately 100 mm/day on July 17, 2006. Taepyung and Samjook display that WRF produced either with KLDAS or with GFS embedded initial soil moisture data higher precipitation amounts compared to observation. Results and discussions in detail on accuracy of prediction using formerly mentioned manners are going to be presented in 2011 Annual Conference of the Korean Society of Hazard Mitigation.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.22 no.1
• /
• pp.135-144
• /
• 2020

Since the depth of tunneling with tunnel boring machine (TBM) becomes deeper and deeper, the expense for site investigation for coring and geophysical survey increases to obtain the sufficient accuracy. The tunnel ahead prediction methods have been introduced to overcome this limitation in the stage of site investigation. Probe drilling can obtain the core and borehole images from a borehole. However, the space in TBM for the probe drilling equipment is restricted and the core from probe drilling cannot reflect the whole tunnel face. Seismic methods such as tunnel seismic prediction (TSP) can forecast over 100 m ahead from the tunnel face though the signal is usually generated using the explosive which can affect the stability of segments and backfill grout. Electromagnetic methods such as tunnel electrical resistivity prospecting system (TEPS) offer the exact prediction for a conductive zone such as water-bearing zone. However, the number of electrodes installed for exploration is limited in small diameter TBM and finally the reduction of prediction ranges. In this study, the theoretical equations for the electrical resistivity survey whose electrodes are installed in the face and side of TBM to minimize the installed electrodes on face. The experimental tests were conducted to verify the derived equations.

• Journal of Navigation and Port Research
• /
• v.36 no.6
• /
• pp.521-526
• /
• 2012

More efficient and effective volume management of trade cargo is recently requested due to FTA with foreign country. Above all, the grain is the main cargo needed in Korean food life and was appointed as the core trade cargo during FTA. This study is aimed to forecast future demands of grain volumes which are handled at Incheon port because most of the grain volumes are traded at Incheon port in Korea. System Dynamics (SD) was used for forecasting as the methodology. Also, population, yearly grain consumption per a man, GDP, GRDP, exchange rate, and BDI were used as the factors that influence grain volumes. Simulation duration was from 2000 to 2020 and real data was used from 2000 to 2007. According to the simulation, 2020's grain volumes at Incheon port were forecasted to be about 2 million tons and grain volumes handled at Incheon port were continuously reduced. In order to measure accuracy of the simulation, this study implemented MAPE analysis. And after the implementation, the simulation was decided as a much more accurate model because MAPE value was calculated to be 6.3%. This study respectively examined factors using the sensitivity analysis. As a result, in terms of the effects on grain volume in Incheon Port, the population factor is most significant and exchange rate factor is the least.

• The KIPS Transactions:PartB
• /
• v.14B no.7
• /
• pp.531-540
• /
• 2007

In this research, we proposed a two-phase hybrid stock price forecasting model with cointegration tests and artificial neural networks. Using not only the related stocks to the target stock but also the past information as input features in neural networks, the new model showed an improved performance in forecasting than that of the usual neural networks. Firstly in order to extract stocks which have long run relationships with the target stock, we made use of Johansen's cointegration test. In stock market, some stocks are apt to vary similarly and these phenomenon can be very informative to forecast the target stock. Johansen's cointegration test provides whether variables are related and whether the relationship is statistically significant. Secondly, we learned the model which includes lagged variables of the target and related stocks in addition to other characteristics of them. Although former research usually did not incorporate those variables, it is well known that most economic time series data are depend on its past value. Also, it is common in econometric literatures to consider lagged values as dependent variables. We implemented a price direction forecasting system for KOSPI index to examine the performance of the proposed model. As the result, our model had 11.29% higher forecasting accuracy on average than the model learned without cointegration test and also showed 10.59% higher on average than the model which randomly selected stocks to make the size of the feature set same as that of the proposed model.