• Journal of Intelligence and Information Systems
• /
• v.21 no.4
• /
• pp.93-110
• /
• 2015

The recommender system is a system which recommends products to the customers who are likely to be interested in. Based on automated information filtering technology, various recommender systems have been developed. Collaborative filtering (CF), one of the most successful recommendation algorithms, has been applied in a number of different domains such as recommending Web pages, books, movies, music and products. But, it has been known that CF has a critical shortcoming. CF finds neighbors whose preferences are like those of the target customer and recommends products those customers have most liked. Thus, CF works properly only when there's a sufficient number of ratings on common product from customers. When there's a shortage of customer ratings, CF makes the formation of a neighborhood inaccurate, thereby resulting in poor recommendations. To improve the performance of CF based recommender systems, most of the related studies have been focused on the development of novel algorithms under the assumption of using a single profile, which is created from user's rating information for items, purchase transactions, or Web access logs. With the advent of big data, companies got to collect more data and to use a variety of information with big size. So, many companies recognize it very importantly to utilize big data because it makes companies to improve their competitiveness and to create new value. In particular, on the rise is the issue of utilizing personal big data in the recommender system. It is why personal big data facilitate more accurate identification of the preferences or behaviors of users. The proposed recommendation methodology is as follows: First, multimodal user profiles are created from personal big data in order to grasp the preferences and behavior of users from various viewpoints. We derive five user profiles based on the personal information such as rating, site preference, demographic, Internet usage, and topic in text. Next, the similarity between users is calculated based on the profiles and then neighbors of users are found from the results. One of three ensemble approaches is applied to calculate the similarity. Each ensemble approach uses the similarity of combined profile, the average similarity of each profile, and the weighted average similarity of each profile, respectively. Finally, the products that people among the neighborhood prefer most to are recommended to the target users. For the experiments, we used the demographic data and a very large volume of Web log transaction for 5,000 panel users of a company that is specialized to analyzing ranks of Web sites. R and SAS E-miner was used to implement the proposed recommender system and to conduct the topic analysis using the keyword search, respectively. To evaluate the recommendation performance, we used 60% of data for training and 40% of data for test. The 5-fold cross validation was also conducted to enhance the reliability of our experiments. A widely used combination metric called F1 metric that gives equal weight to both recall and precision was employed for our evaluation. As the results of evaluation, the proposed methodology achieved the significant improvement over the single profile based CF algorithm. In particular, the ensemble approach using weighted average similarity shows the highest performance. That is, the rate of improvement in F1 is 16.9 percent for the ensemble approach using weighted average similarity and 8.1 percent for the ensemble approach using average similarity of each profile. From these results, we conclude that the multimodal profile ensemble approach is a viable solution to the problems encountered when there's a shortage of customer ratings. This study has significance in suggesting what kind of information could we use to create profile in the environment of big data and how could we combine and utilize them effectively. However, our methodology should be further studied to consider for its real-world application. We need to compare the differences in recommendation accuracy by applying the proposed method to different recommendation algorithms and then to identify which combination of them would show the best performance.

• Asian Pacific Journal of Cancer Prevention
• /
• v.14 no.12
• /
• pp.7509-7515
• /
• 2013

To assess the risk of cancers associated with sleep duration using meta-analysis of published cohort studies, we performed a comprehensive search using PubMed, Embase and Web of Science through October 2013. We combined hazard ratios (HRs) from individual studies using meta-analysis approaches. A random effect dose-response analysis was used to evaluate the relationship between sleep duration and cancer risk. Subgroup analyses and sensitivity analyses were also performed. Publication bias was evaluated using Funnel plots and Begg's test. A total of 13 cohorts from 12 studies were included in this meta-analysis, which included 723, 337 participants with 15, 156 reported cancer outcomes during a follow-up period ranging from 7.5 to 22 years. The pooled adjusted HRs were 1.06 (95% CI: 0.92, 1.23; P for heterogeneity =0.003) for short sleep duration, 0.91 (95% CI: 0.78, 1.07; P for heterogeneity <0.0001) for long sleep duration. In subgroup analyses stratified by cancer type, long duration of sleep showed an inverse relation with hormone-related cancer (HR=0.79; 95% CI: 0.65, 0.97; P for heterogeneity =0.009) and a greater risk of colorectal cancer (HR=1.29; 95% CI: 1.09, 1.52; P for heterogeneity =0.346). Further meta-analysis on dose-response relationships showed that the relative risks of cancer were 1.00 (95% CI: 0.99, 1.01; P for linear trend=0.9151) for one hour of sleep increment per day, and 1.00 (95% CI: 0.98, 1.01; P for linear trend=0.7749) for one hour of sleep increment per night. No significant dose-response relationship between sleep duration and cancer was found on non-linearity testing (P=0.5053). Our meta-analysis suggests a positive association between long sleep duration and colorectal cancer, and an inverse association with incidence of hormone related cancers like those in the breast. Studies with larger sample size, longer follow-up times, more cancer types and detailed measure of sleep duration are warranted to confirm these results.

• Journal of Plant Biotechnology
• /
• v.42 no.1
• /
• pp.60-70
• /
• 2015

The aim of this study was to investigate whether fourier transform infrared (FT-IR) spectroscopy can be applied to simultaneous determination of fatty acids contents in different soybean cultivars. Total 153 lines of soybean (Glycine max Merrill) were examined by FT-IR spectroscopy. Quantification of fatty acids from the soybean lines was confirmed by quantitative gas chromatography (GC) analysis. The quantitative spectral variation among different soybean lines was observed in the amide bond region ($1,700{\sim}1,500cm^{-1}$), phosphodiester groups ($1,500{\sim}1,300cm^{-1}$) and sugar region ($1,200{\sim}1,000cm^{-1}$) of FT-IR spectra. The quantitative prediction modeling of 5 individual fatty acids contents (palmitic acid, stearic acid, oleic acid, linoleic acid, linolenic acid) from soybean lines were established using partial least square regression algorithm from FT-IR spectra. In cross validation, there were high correlations ($R^2{\geq}0.97$) between predicted content of 5 individual fatty acids by PLS regression modeling from FT-IR spectra and measured content by GC. In external validation, palmitic acid ($R^2=0.8002$), oleic acid ($R^2=0.8909$) and linoleic acid ($R^2=0.815$) were predicted with good accuracy, while prediction for stearic acid ($R^2=0.4598$), linolenic acid ($R^2=0.6868$) had relatively lower accuracy. These results clearly show that FT-IR spectra combined with multivariate analysis can be used to accurately predict fatty acids contents in soybean lines. Therefore, we suggest that the PLS prediction system for fatty acid contents using FT-IR analysis could be applied as a rapid and high throughput screening tool for the breeding for modified Fatty acid composition in soybean and contribute to accelerating the conventional breeding.

• Journal of Intelligence and Information Systems
• /
• v.27 no.3
• /
• pp.157-173
• /
• 2021

With the development of information technology, the amount of available information increases daily. However, having access to so much information makes it difficult for users to easily find the information they seek. Users want a visualized system that reduces information retrieval and learning time, saving them from personally reading and judging all available information. As a result, recommendation systems are an increasingly important technologies that are essential to the business. Collaborative filtering is used in various fields with excellent performance because recommendations are made based on similar user interests and preferences. However, limitations do exist. Sparsity occurs when user-item preference information is insufficient, and is the main limitation of collaborative filtering. The evaluation value of the user item matrix may be distorted by the data depending on the popularity of the product, or there may be new users who have not yet evaluated the value. The lack of historical data to identify consumer preferences is referred to as data sparsity, and various methods have been studied to address these problems. However, most attempts to solve the sparsity problem are not optimal because they can only be applied when additional data such as users' personal information, social networks, or characteristics of items are included. Another problem is that real-world score data are mostly biased to high scores, resulting in severe imbalances. One cause of this imbalance distribution is the purchasing bias, in which only users with high product ratings purchase products, so those with low ratings are less likely to purchase products and thus do not leave negative product reviews. Due to these characteristics, unlike most users' actual preferences, reviews by users who purchase products are more likely to be positive. Therefore, the actual rating data is over-learned in many classes with high incidence due to its biased characteristics, distorting the market. Applying collaborative filtering to these imbalanced data leads to poor recommendation performance due to excessive learning of biased classes. Traditional oversampling techniques to address this problem are likely to cause overfitting because they repeat the same data, which acts as noise in learning, reducing recommendation performance. In addition, pre-processing methods for most existing data imbalance problems are designed and used for binary classes. Binary class imbalance techniques are difficult to apply to multi-class problems because they cannot model multi-class problems, such as objects at cross-class boundaries or objects overlapping multiple classes. To solve this problem, research has been conducted to convert and apply multi-class problems to binary class problems. However, simplification of multi-class problems can cause potential classification errors when combined with the results of classifiers learned from other sub-problems, resulting in loss of important information about relationships beyond the selected items. Therefore, it is necessary to develop more effective methods to address multi-class imbalance problems. We propose a collaborative filtering model using CGAN to generate realistic virtual data to populate the empty user-item matrix. Conditional vector y identify distributions for minority classes and generate data reflecting their characteristics. Collaborative filtering then maximizes the performance of the recommendation system via hyperparameter tuning. This process should improve the accuracy of the model by addressing the sparsity problem of collaborative filtering implementations while mitigating data imbalances arising from real data. Our model has superior recommendation performance over existing oversampling techniques and existing real-world data with data sparsity. SMOTE, Borderline SMOTE, SVM-SMOTE, ADASYN, and GAN were used as comparative models and we demonstrate the highest prediction accuracy on the RMSE and MAE evaluation scales. Through this study, oversampling based on deep learning will be able to further refine the performance of recommendation systems using actual data and be used to build business recommendation systems.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.4
• /
• pp.189-198
• /
• 2024

Lipreading is one of the important parts of speech recognition, and several studies have been conducted to improve the performance of lipreading in lipreading systems for speech recognition. Recent studies have used method to modify the model architecture of lipreading system to improve recognition performance. Unlike previous research that improve recognition performance by modifying model architecture, we aim to improve recognition performance without any change in model architecture. In order to improve the recognition performance without modifying the model architecture, we refer to the cues used in human lipreading and set other regions such as chin and cheeks as regions of interest along with the lip region, which is the existing region of interest of lipreading systems, and compare the recognition rate of each region of interest to propose the highest performing region of interest In addition, assuming that the difference in normalization results caused by the difference in interpolation method during the process of normalizing the size of the region of interest affects the recognition performance, we interpolate the same region of interest using nearest neighbor interpolation, bilinear interpolation, and bicubic interpolation, and compare the recognition rate of each interpolation method to propose the best performing interpolation method. Each region of interest was detected by training an object detection neural network, and dynamic time warping templates were generated by normalizing each region of interest, extracting and combining features, and mapping the dimensionality reduction of the combined features into a low-dimensional space. The recognition rate was evaluated by comparing the distance between the generated dynamic time warping templates and the data mapped to the low-dimensional space. In the comparison of regions of interest, the result of the region of interest containing only the lip region showed an average recognition rate of 97.36%, which is 3.44% higher than the average recognition rate of 93.92% in the previous study, and in the comparison of interpolation methods, the bilinear interpolation method performed 97.36%, which is 14.65% higher than the nearest neighbor interpolation method and 5.55% higher than the bicubic interpolation method. The code used in this study can be found a https://github.com/haraisi2/Lipreading-Systems.

• Journal of National Security and Military Science
• /
• s.8
• /
• pp.49-168
• /
• 2010

Just before the Korean War, the total number of the North Korean troops was 198,380, while that of the ROK(Republic of Korea) army troops 105,752. That is, the total number of the ROK army troops at that time was 53.3% of the total number of the North Korean army. As of December 2008, the total number of the North Korean troops is estimated to be 1,190,000, while that of the ROK troops is 655,000, so the ROK army maintains 55.04% of the total number of the North Korean troops. If the ROK army continues to reduce its troops according to [Military Reform Plan 2020], the total number of its troops will be 517,000 m 2020. If North Korea maintains the current status(l,190,000 troops), the number of the ROK troops will be 43.4% of the North Korean army. In terms of units, just before the Korean War, the number of the ROK army divisions and regiments was 80% and 44.8% of North Korean army. As of December 2008, North Korea maintains 86 divisions and 69 regiments. Compared to the North Korean army, the ROK army maintains 46 Divisions (53.4% of North Korean army) and 15 regiments (21.3% of North Korean army). If the ROK army continue to reduce the military units according to [Military Reform Plan 2020], the number of ROK army divisions will be 28(13 Active Division, 4 Mobilization Divisions and 11 Local Reserve Divisions), while that of the North Korean army will be 86 in 2020. In that case, the number of divisions of the ROK army will be 32.5% of North Korean army. During the Korean war, North Korea suddenly invaded the Republic of Korea and occupied its capital 3 days after the war began. At that time, the ROK army maintained 80% of army divisions, compared to the North Korean army. The lesson to be learned from this is that, if the ROK army is forced to disperse its divisions because of the simultaneous invasion of North Korea and attack of guerrillas in home front areas, the Republic of Korea can be in a serious military danger, even though it maintains 80% of military divisions of North Korea. If the ROK army promotes the plans in [Military Reform Plan 2020], the number of military units of the ROK army will be 32.5% of that of the North Korean army. This ratio is 2.4 times lower than that of the time when the Korean war began, and in this case, 90% of total military power should be placed in the DMZ area. If 90% of military power is placed in the DMZ area, few troops will be left for the defense of home front. In addition, if the ROK army continues to reduce the troops, it can allow North Korea to have asymmetrical superiority in military force and it will eventually exert negative influence on the stability and peace of the Korean peninsular. On the other hand, it should be reminded that, during the Korean War, the Republic of Korea was attacked by North Korea, though it kept 53.3% of troops, compared to North Korea. It should also be reminded that, as of 2008, the ROK army is defending its territory with the troops 55.04% of North Korea. Moreover, the national defense is assisted by 25,120 troops of the US Forces in Korea. In case the total number of the ROK troops falls below 43.4% of the North Korean army, it may cause social unrest about the national security and may lead North Korea's misjudgement. Besides, according to Lanchester strategy, the party with weaker military power (60% compared to the party with stronger military power) has the 4.1% of winning possibility. Therefore, if we consider the fact that the total number of the ROK army troops is 55.04% of that of the North Korean army, the winning possibility of the ROK army is not higher than 4.1%. If the total number of ROK troops is reduced to 43.4% of that of North Korea, the winning possibility will be lower and the military operations will be in critically difficult situation. [Military Reform Plan 2020] rums at the reduction of troops and units of the ground forces under the policy of 'select few'. However, the problem is that the financial support to achieve this goal is not secured. Therefore, the promotion of [Military Reform Plan 2020] may cause the weakening of military defence power in 2020. Some advanced countries such as Japan, UK, Germany, and France have promoted the policy of 'select few'. However, what is to be noted is that the national security situation of those countries is much different from that of Korea. With the collapse of the Soviet Unions and European communist countries, the military threat of those European advanced countries has almost disappeared. In addition, the threats those advanced countries are facing are not wars in national level, but terrorism in international level. To cope with the threats like terrorism, large scaled army trops would not be necessary. So those advanced European countries can promote the policy of 'select few'. In line with this, those European countries put their focuses on the development of military sections that deal with non-military operations and protection from unspecified enemies. That is, those countries are promoting the policy of 'select few', because they found that the policy is suitable for their national security environment. Moreover, since they are pursuing common interest under the European Union(EU) and they can form an allied force under NATO, it is natural that they are pursing the 'select few' policy. At present, NATO maintains the larger number of troops(2,446,000) than Russia(l,027,000) to prepare for the potential threat of Russia. The situation of japan is also much different from that of Korea. As a country composed of islands, its prime military focus is put on the maritime defense. Accordingly, the development of ground force is given secondary focus. The japanese government promotes the policy to develop technology-concentrated small size navy and air-forces, instead of maintaining large-scaled ground force. In addition, because of the 'Peace Constitution' that was enacted just after the end of World War II, japan cannot maintain troops more than 240,000. With the limited number of troops (240,000), japan has no choice but to promote the policy of 'select few'. However, the situation of Korea is much different from the situations of those countries. The Republic of Korea is facing the threat of the North Korean Army that aims at keeping a large-scale military force. In addition, the countries surrounding Korea are also super powers containing strong military forces. Therefore, to cope with the actual threat of present and unspecified threat of future, the importance of maintaining a carefully calculated large-scale military force cannot be denied. Furthermore, when considering the fact that Korea is in a peninsular, the Republic of Korea must take it into consideration the tradition of continental countries' to maintain large-scale military powers. Since the Korean War, the ROK army has developed the technology-force combined military system, maintaining proper number of troops and units and pursuing 'select few' policy at the same time. This has been promoted with the consideration of military situation in the Koran peninsular and the cooperation of ROK-US combined forces. This kind of unique military system that cannot be found in other countries can be said to be an insightful one for the preparation for the actual threat of North Korea and the conflicts between continental countries and maritime countries. In addition, this kind of technology-force combined military system has enabled us to keep peace in Korea. Therefore, it would be desirable to maintain this technology-force combined military system until the reunification of the Korean peninsular. Furthermore, it is to be pointed out that blindly following the 'select few' policy of advanced countries is not a good option, because it is ignoring the military strategic situation of the Korean peninsular. If the Republic of Korea pursues the reduction of troops and units radically without consideration of the threat of North Korea and surrounding countries, it could be a significant strategic mistake. In addition, the ROK army should keep an eye on the fact the European advanced countries and Japan that are not facing direct military threats are spending more defense expenditures than Korea. If the ROK army reduces military power without proper alternatives, it would exert a negative effect on the stable economic development of Korea and peaceful reunification of the Korean peninsular. Therefore, the desirable option would be to focus on the development of quality of forces, maintaining proper size and number of troops and units under the technology-force combined military system. The tableau above shows that the advanced countries like the UK, Germany, Italy, and Austria spend more defense expenditure per person than the Republic of Korea, although they do not face actual military threats, and that they keep achieving better economic progress than the countries that spend less defense expenditure. Therefore, it would be necessary to adopt the merits of the defense systems of those advanced countries. As we have examined, it would be desirable to maintain the current size and number of troops and units, to promote 'select few' policy with increased defense expenditure, and to strengthen the technology-force combined military system. On the basis of firm national security, the Republic of Korea can develop efficient policies for reunification and prosperity, and jump into the status of advanced countries. Therefore, the plans to reduce troops and units in [Military Reform Plan 2020] should be reexamined. If it is difficult for the ROK army to maintain its size of 655,000 troops because of low birth rate, the plans to establish the prompt mobilization force or to adopt drafting system should be considered for the maintenance of proper number of troops and units. From now on, the Republic of Korean government should develop plans to keep peace as well as to prepare unexpected changes in the Korean peninsular. For the achievement of these missions, some options can be considered. The first one is to maintain the same size of military troops and units as North Korea. The second one is to maintain the same level of military power as North Korea in terms of military force index. The third one is to maintain the same level of military power as North Korea, with the combination of the prompt mobilization force and the troops in active service under the system of technology-force combined military system. At present, it would be not possible for the ROK army to maintain such a large-size military force as North Korea (1,190,000 troops and 86 units). So it would be rational to maintain almost the same level of military force as North Korea with the combination of the troops on the active list and the prompt mobilization forces. In other words, with the combination of the troops in active service (60%) and the prompt mobilization force (40%), the ROK army should develop the strategies to harmonize technology and forces. The Korean government should also be prepared for the strategic flexibility of USFK, the possibility of American policy change about the location of foreign army, radical unexpected changes in North Korea, the emergence of potential threat, surrounding countries' demand for Korean force for the maintenance of regional stability, and demand for international cooperation against terrorism. For this, it is necessary to develop new approaches toward the proper number and size of troops and units. For instance, to prepare for radical unexpected political or military changes in North Korea, the Republic of Korea should have plans to protect a large number of refugees, to control arms and people, to maintain social security, and to keep orders in North Korea. From the experiences of other countries, it is estimated that 115,000 to 230,000 troops, plus ten thousands of police are required to stabilize the North Korean society, in the case radical unexpected military or political change happens in North Korea. In addition, if the Republic of Korea should perform the release of hostages, control of mass destruction weapons, and suppress the internal wars in North Korea, it should send 460,000 troops to North Korea. Moreover, if the Republic of Korea wants to stop the attack of North Korea and flow of refugees in DMZ area, at least 600,000 troops would be required. In sum, even if the ROK army maintains 600,000 troops, it may need additional 460,000 troops to prepare for unexpected radical changes in North Korea. For this, it is necessary to establish the prompt mobilization force whose size and number are almost the same as the troops in active service. In case the ROK army keeps 650,000 troops, the proper number of the prompt mobilization force would be 460,000 to 500,000.

• Journal of Intelligence and Information Systems
• /
• v.25 no.2
• /
• pp.39-55
• /
• 2019

Recently banks and large financial institutions have introduced lots of Robo-Advisor products. Robo-Advisor is a Robot to produce the optimal asset allocation portfolio for investors by using the financial engineering algorithms without any human intervention. Since the first introduction in Wall Street in 2008, the market size has grown to 60 billion dollars and is expected to expand to 2,000 billion dollars by 2020. Since Robo-Advisor algorithms suggest asset allocation output to investors, mathematical or statistical asset allocation strategies are applied. Mean variance optimization model developed by Markowitz is the typical asset allocation model. The model is a simple but quite intuitive portfolio strategy. For example, assets are allocated in order to minimize the risk on the portfolio while maximizing the expected return on the portfolio using optimization techniques. Despite its theoretical background, both academics and practitioners find that the standard mean variance optimization portfolio is very sensitive to the expected returns calculated by past price data. Corner solutions are often found to be allocated only to a few assets. The Black-Litterman Optimization model overcomes these problems by choosing a neutral Capital Asset Pricing Model equilibrium point. Implied equilibrium returns of each asset are derived from equilibrium market portfolio through reverse optimization. The Black-Litterman model uses a Bayesian approach to combine the subjective views on the price forecast of one or more assets with implied equilibrium returns, resulting a new estimates of risk and expected returns. These new estimates can produce optimal portfolio by the well-known Markowitz mean-variance optimization algorithm. If the investor does not have any views on his asset classes, the Black-Litterman optimization model produce the same portfolio as the market portfolio. What if the subjective views are incorrect? A survey on reports of stocks performance recommended by securities analysts show very poor results. Therefore the incorrect views combined with implied equilibrium returns may produce very poor portfolio output to the Black-Litterman model users. This paper suggests an objective investor views model based on Support Vector Machines(SVM), which have showed good performance results in stock price forecasting. SVM is a discriminative classifier defined by a separating hyper plane. The linear, radial basis and polynomial kernel functions are used to learn the hyper planes. Input variables for the SVM are returns, standard deviations, Stochastics %K and price parity degree for each asset class. SVM output returns expected stock price movements and their probabilities, which are used as input variables in the intelligent views model. The stock price movements are categorized by three phases; down, neutral and up. The expected stock returns make P matrix and their probability results are used in Q matrix. Implied equilibrium returns vector is combined with the intelligent views matrix, resulting the Black-Litterman optimal portfolio. For comparisons, Markowitz mean-variance optimization model and risk parity model are used. The value weighted market portfolio and equal weighted market portfolio are used as benchmark indexes. We collect the 8 KOSPI 200 sector indexes from January 2008 to December 2018 including 132 monthly index values. Training period is from 2008 to 2015 and testing period is from 2016 to 2018. Our suggested intelligent view model combined with implied equilibrium returns produced the optimal Black-Litterman portfolio. The out of sample period portfolio showed better performance compared with the well-known Markowitz mean-variance optimization portfolio, risk parity portfolio and market portfolio. The total return from 3 year-period Black-Litterman portfolio records 6.4%, which is the highest value. The maximum draw down is -20.8%, which is also the lowest value. Sharpe Ratio shows the highest value, 0.17. It measures the return to risk ratio. Overall, our suggested view model shows the possibility of replacing subjective analysts's views with objective view model for practitioners to apply the Robo-Advisor asset allocation algorithms in the real trading fields.

• Management & Information Systems Review
• /
• v.39 no.2
• /
• pp.39-60
• /
• 2020

Block chain technology is one of the core elements for realizing the 4^th industrial revolution, and many efforts have been made by government and companies to provide services based on block chain technology. In this study we analyzed the benefits of block chain technology for EVMS and designed EVMS block chain platform with increased data security and work efficiency for project management data, which are important assets in monitoring progress, foreseeing future events, and managing post-completion. We did the case studies on the benefits of block chain technology and then conducted the survey study on security, reliability, and efficiency of block chain technology, targeting 18 block chain experts and project developers. And then, we interviewed EVMS system operator on the compatibility between block chain technology and EVM Systems. The result of the case studies showed that block chain technology can be applied to financial, logistic, medical, and public services to simplify the insurance claim process and to improve reliability by distributing transaction data storage and applying security·encryption features. Also, our research on the characteristics and necessity of block chain technology in EVMS revealed the improvability of security, reliability, and efficiency of management and distribution of EVMS data. Finally, we designed a network model, a block structure, and a consensus algorithm model and combined them to construct a conceptual block chain model for EVM system. This study has the following contribution. First, we reviewed that the block chain technology is suitable for application in the defense sector and proposed a conceptual model. Second, the effect that can be obtained by applying block chain technology to EVMS was derived, and the possibility of improving the existing business process was derived.

• Journal of Intelligence and Information Systems
• /
• v.21 no.2
• /
• pp.1-18
• /
• 2015

Collaborative filtering(CF) algorithm has been popularly used for recommender systems in both academic and practical applications. A general CF system compares users based on how similar they are, and creates recommendation results with the items favored by other people with similar tastes. Thus, it is very important for CF to measure the similarities between users because the recommendation quality depends on it. In most cases, users' explicit numeric ratings of items(i.e. quantitative information) have only been used to calculate the similarities between users in CF. However, several studies indicated that qualitative information such as user's reviews on the items may contribute to measure these similarities more accurately. Considering that a lot of people are likely to share their honest opinion on the items they purchased recently due to the advent of the Web 2.0, user's reviews can be regarded as the informative source for identifying user's preference with accuracy. Under this background, this study proposes a new hybrid recommender system that combines with users' review mining. Our proposed system is based on conventional memory-based CF, but it is designed to use both user's numeric ratings and his/her text reviews on the items when calculating similarities between users. In specific, our system creates not only user-item rating matrix, but also user-item review term matrix. Then, it calculates rating similarity and review similarity from each matrix, and calculates the final user-to-user similarity based on these two similarities(i.e. rating and review similarities). As the methods for calculating review similarity between users, we proposed two alternatives - one is to use the frequency of the commonly used terms, and the other one is to use the sum of the importance weights of the commonly used terms in users' review. In the case of the importance weights of terms, we proposed the use of average TF-IDF(Term Frequency - Inverse Document Frequency) weights. To validate the applicability of the proposed system, we applied it to the implementation of a recommender system for smartphone applications (hereafter, app). At present, over a million apps are offered in each app stores operated by Google and Apple. Due to this information overload, users have difficulty in selecting proper apps that they really want. Furthermore, app store operators like Google and Apple have cumulated huge amount of users' reviews on apps until now. Thus, we chose smartphone app stores as the application domain of our system. In order to collect the experimental data set, we built and operated a Web-based data collection system for about two weeks. As a result, we could obtain 1,246 valid responses(ratings and reviews) from 78 users. The experimental system was implemented using Microsoft Visual Basic for Applications(VBA) and SAS Text Miner. And, to avoid distortion due to human intervention, we did not adopt any refining works by human during the user's review mining process. To examine the effectiveness of the proposed system, we compared its performance to the performance of conventional CF system. The performances of recommender systems were evaluated by using average MAE(mean absolute error). The experimental results showed that our proposed system(MAE = 0.7867 ~ 0.7881) slightly outperformed a conventional CF system(MAE = 0.7939). Also, they showed that the calculation of review similarity between users based on the TF-IDF weights(MAE = 0.7867) leaded to better recommendation accuracy than the calculation based on the frequency of the commonly used terms in reviews(MAE = 0.7881). The results from paired samples t-test presented that our proposed system with review similarity calculation using the frequency of the commonly used terms outperformed conventional CF system with 10% statistical significance level. Our study sheds a light on the application of users' review information for facilitating electronic commerce by recommending proper items to users.

• Proceedings of the Korean Powder Metallurgy Institute Conference
• /
• 2000.04a
• /
• pp.9-10
• /
• 2000

An important business-field of world-wide steel-industry is the coating of thin metal-sheets with zinc, zinc-aluminum and aluminum based materials. These products mostly go into automotive industry. in particular for the car-body. into building and construction industry as well as household appliances. Due to mass-production, the processing is done in large continuously operating plants where the mostly cold-rolled metal-strip as the substrate is handled in coils up to 40 tons unwind before and rolled up again after passing the processing plant which includes cleaning, annealing, hot-dip galvanizing / aluminizing and chemical treatment. In the liquid Zn, Zn-AI, AI-Zn and AI-Si bathes a combined action of corrosion and wear under high temperature and high stress onto the transfer components (rolls) accounts for major economic losses. Most critical here are the bearing systems of these rolls operating in the liquid system. Rolls in liquid system can not be avoided as they are needed to transfer the steel-strip into and out of the crucible. Since several years, ceramic roller bearings are tested here [1.2], however, in particular due to uncontrollable Slag-impurities within the hot bath [3], slide bearings are still expected to be of a higher potential [4]. The today's state of the art is the application of slide bearings based on Stellite\ulcorneragainst Stellite which is in general a 50-60 wt% Co-matrix with incorporated Cr- and W-carbides and other composites. Indeed Stellite is used as the bearing-material as of it's chemical properties (does not go into solution), the physical properties in particular with poor lubricating properties are not satisfying at all. To increase the Sliding behavior in the bearing system, about 0.15-0.2 wt% of lead has been added into the hot-bath in the past. Due to environmental regulations. this had to be reduced dramatically_ This together with the heavily increasing production rates expressed by increased velocity of the substrate-steel-band up to 200 m/min and increased tractate power up to 10 tons in modern plants. leads to life times of the bearings of a few up to several days only. To improve this situation. the Mechanical Alloying (MA) TeChnique [5.6.7.8] is used to prOduce advanced Stellite-based bearing materials. A lubricating phase is introduced into Stellite-powder-material by MA, the composite-powder-particles are coated by High Energy Milling (HEM) in order to produce bearing-bushes of approximately 12 kg by Sintering, Liquid Phase Sintering (LPS) and Hot Isostatic Pressing (HIP). The chemical and physical behavior of samples as well as the bearing systems in the hot galvanizing / aluminizing plant are discussed. DependenCies like lubricant material and composite, LPS-binder and composite, particle shape and PM-route with respect to achievable density. (temperature--) shock-reSistibility and corrosive-wear behavior will be described. The materials are characterized by particle size analysis (laser diffraction), scanning electron microscopy and X-ray diffraction. corrosive-wear behavior is determined using a special cylinder-in-bush apparatus (CIBA) as well as field-test in real production condition. Part I of this work describes the initial testing phase where different sample materials are produced, characterized, consolidated and tested in the CIBA under a common AI-Zn-system. The results are discussed and the material-system for the large components to be produced for the field test in real production condition is decided. Outlook: Part II of this work will describe the field test in a hot-dip-galvanizing/aluminizing plant of the mechanically alloyed bearing bushes under aluminum-rich liquid metal. Alter testing, the bushes will be characterized and obtained results with respect to wear. expected lifetime, surface roughness and infiltration will be discussed. Part III of this project will describe a second initial testing phase where the won results of part 1+11 will be transferred to the AI-Si system. Part IV of this project will describe the field test in a hot-dip-aluminizing plant of the mechanically alloyed bearing bushes under aluminum liquid metal. After testing. the bushes will be characterized and obtained results with respect to wear. expected lifetime, surface roughness and infiltration will be discussed.