• Information Systems Review
• /
• v.16 no.3
• /
• pp.161-177
• /
• 2014

Corporate credit rating assessment consists of complicated processes in which various factors describing a company are taken into consideration. Such assessment is known to be very expensive since domain experts should be employed to assess the ratings. As a result, the data-driven corporate credit rating prediction using statistical and artificial intelligence (AI) techniques has received considerable attention from researchers and practitioners. In particular, statistical methods such as multiple discriminant analysis (MDA) and multinomial logistic regression analysis (MLOGIT), and AI methods including case-based reasoning (CBR), artificial neural network (ANN), and multiclass support vector machine (MSVM) have been applied to corporate credit rating.2) Among them, MSVM has recently become popular because of its robustness and high prediction accuracy. In this study, we propose a novel optimized MSVM model, and appy it to corporate credit rating prediction in order to enhance the accuracy. Our model, named 'GAMSVM (Genetic Algorithm-optimized Multiclass Support Vector Machine),' is designed to simultaneously optimize the kernel parameters and the feature subset selection. Prior studies like Lorena and de Carvalho (2008), and Chatterjee (2013) show that proper kernel parameters may improve the performance of MSVMs. Also, the results from the studies such as Shieh and Yang (2008) and Chatterjee (2013) imply that appropriate feature selection may lead to higher prediction accuracy. Based on these prior studies, we propose to apply GAMSVM to corporate credit rating prediction. As a tool for optimizing the kernel parameters and the feature subset selection, we suggest genetic algorithm (GA). GA is known as an efficient and effective search method that attempts to simulate the biological evolution phenomenon. By applying genetic operations such as selection, crossover, and mutation, it is designed to gradually improve the search results. Especially, mutation operator prevents GA from falling into the local optima, thus we can find the globally optimal or near-optimal solution using it. GA has popularly been applied to search optimal parameters or feature subset selections of AI techniques including MSVM. With these reasons, we also adopt GA as an optimization tool. To empirically validate the usefulness of GAMSVM, we applied it to a real-world case of credit rating in Korea. Our application is in bond rating, which is the most frequently studied area of credit rating for specific debt issues or other financial obligations. The experimental dataset was collected from a large credit rating company in South Korea. It contained 39 financial ratios of 1,295 companies in the manufacturing industry, and their credit ratings. Using various statistical methods including the one-way ANOVA and the stepwise MDA, we selected 14 financial ratios as the candidate independent variables. The dependent variable, i.e. credit rating, was labeled as four classes: 1(A1); 2(A2); 3(A3); 4(B and C). 80 percent of total data for each class was used for training, and remaining 20 percent was used for validation. And, to overcome small sample size, we applied five-fold cross validation to our dataset. In order to examine the competitiveness of the proposed model, we also experimented several comparative models including MDA, MLOGIT, CBR, ANN and MSVM. In case of MSVM, we adopted One-Against-One (OAO) and DAGSVM (Directed Acyclic Graph SVM) approaches because they are known to be the most accurate approaches among various MSVM approaches. GAMSVM was implemented using LIBSVM-an open-source software, and Evolver 5.5-a commercial software enables GA. Other comparative models were experimented using various statistical and AI packages such as SPSS for Windows, Neuroshell, and Microsoft Excel VBA (Visual Basic for Applications). Experimental results showed that the proposed model-GAMSVM-outperformed all the competitive models. In addition, the model was found to use less independent variables, but to show higher accuracy. In our experiments, five variables such as X7 (total debt), X9 (sales per employee), X13 (years after founded), X15 (accumulated earning to total asset), and X39 (the index related to the cash flows from operating activity) were found to be the most important factors in predicting the corporate credit ratings. However, the values of the finally selected kernel parameters were found to be almost same among the data subsets. To examine whether the predictive performance of GAMSVM was significantly greater than those of other models, we used the McNemar test. As a result, we found that GAMSVM was better than MDA, MLOGIT, CBR, and ANN at the 1% significance level, and better than OAO and DAGSVM at the 5% significance level.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.7 no.1
• /
• pp.861-876
• /
• 1965

During my stay in the Netherlands, I have studied the following, primarily in relation to the Mokpo Yong-san project which had been studied by the NEDECO for a feasibility report. 1. Unit hydrograph at Naju There are many ways to make unit hydrograph, but I want explain here to make unit hydrograph from the- actual run of curve at Naju. A discharge curve made from one rain storm depends on rainfall intensity per houre After finriing hydrograph every two hours, we will get two-hour unit hydrograph to devide each ordinate of the two-hour hydrograph by the rainfall intensity. I have used one storm from June 24 to June 26, 1963, recording a rainfall intensity of average 9. 4 mm per hour for 12 hours. If several rain gage stations had already been established in the catchment area. above Naju prior to this storm, I could have gathered accurate data on rainfall intensity throughout the catchment area. As it was, I used I the automatic rain gage record of the Mokpo I moteorological station to determine the rainfall lntensity. In order. to develop the unit ~Ydrograph at Naju, I subtracted the basic flow from the total runoff flow. I also tried to keed the difference between the calculated discharge amount and the measured discharge less than 1O~ The discharge period. of an unit graph depends on the length of the catchment area. 2. Determination of sluice dimension Acoording to principles of design presently used in our country, a one-day storm with a frequency of 20 years must be discharged in 8 hours. These design criteria are not adequate, and several dams have washed out in the past years. The design of the spillway and sluice dimensions must be based on the maximun peak discharge flowing into the reservoir to avoid crop and structure damages. The total flow into the reservoir is the summation of flow described by the Mokpo hydrograph, the basic flow from all the catchment areas and the rainfall on the reservoir area. To calculate the amount of water discharged through the sluiceCper half hour), the average head during that interval must be known. This can be calculated from the known water level outside the sluiceCdetermined by the tide) and from an estimated water level inside the reservoir at the end of each time interval. The total amount of water discharged through the sluice can be calculated from this average head, the time interval and the cross-sectional area of' the sluice. From the inflow into the .reservoir and the outflow through the sluice gates I calculated the change in the volume of water stored in the reservoir at half-hour intervals. From the stored volume of water and the known storage capacity of the reservoir, I was able to calculate the water level in the reservoir. The Calculated water level in the reservoir must be the same as the estimated water level. Mean stand tide will be adequate to use for determining the sluice dimension because spring tide is worse case and neap tide is best condition for the I result of the calculatio 3. Tidal computation for determination of the closure curve. During the construction of a dam, whether by building up of a succession of horizontael layers or by building in from both sides, the velocity of the water flowinii through the closing gapwill increase, because of the gradual decrease in the cross sectional area of the gap. 1 calculated the . velocities in the closing gap during flood and ebb for the first mentioned method of construction until the cross-sectional area has been reduced to about 25% of the original area, the change in tidal movement within the reservoir being negligible. Up to that point, the increase of the velocity is more or less hyperbolic. During the closing of the last 25 % of the gap, less water can flow out of the reservoir. This causes a rise of the mean water level of the reservoir. The difference in hydraulic head is then no longer negligible and must be taken into account. When, during the course of construction. the submerged weir become a free weir the critical flow occurs. The critical flow is that point, during either ebb or flood, at which the velocity reaches a maximum. When the dam is raised further. the velocity decreases because of the decrease\ulcorner in the height of the water above the weir. The calculation of the currents and velocities for a stage in the closure of the final gap is done in the following manner; Using an average tide with a neglible daily quantity, I estimated the water level on the pustream side of. the dam (inner water level). I determined the current through the gap for each hour by multiplying the storage area by the increment of the rise in water level. The velocity at a given moment can be determined from the calcalated current in m3/sec, and the cross-sectional area at that moment. At the same time from the difference between inner water level and tidal level (outer water level) the velocity can be calculated with the formula $h= \frac{V^2}{2g}$ and must be equal to the velocity detertnined from the current. If there is a difference in velocity, a new estimate of the inner water level must be made and entire procedure should be repeated. When the higher water level is equal to or more than 2/3 times the difference between the lower water level and the crest of the dam, we speak of a "free weir." The flow over the weir is then dependent upon the higher water level and not on the difference between high and low water levels. When the weir is "submerged", that is, the higher water level is less than 2/3 times the difference between the lower water and the crest of the dam, the difference between the high and low levels being decisive. The free weir normally occurs first during ebb, and is due to. the fact that mean level in the estuary is higher than the mean level of . the tide in building dams with barges the maximum velocity in the closing gap may not be more than 3m/sec. As the maximum velocities are higher than this limit we must use other construction methods in closing the gap. This can be done by dump-cars from each side or by using a cable way.e or by using a cable way.

• The Korean Journal of Nuclear Medicine Technology
• /
• v.21 no.1
• /
• pp.54-59
• /
• 2017

Purpose The present study aimed at assessing the effectiveness of the respiratory gating method used in the flow mode and additional localized respiratory-gated imaging, which differs from the step and go method. Materials and Methods Respiratory gated imaging was performed in the flow mode to twenty patients with lung cancer (10 patients with stable signals and 10 patients with unstable signals), who underwent PET/CT scanning of the torso using Biograph mCT Flow PET/CT at Bundang Seoul University Hospital from June 2016 to September 2016. Additional images of the lungs were obtained by using the respiratory gating method. SUVmax, SUVmean, and Tumor Volume ($cm^3$) of non-gating images, gating images, and additional lung gating images were found with Syngo,bia (Siemens, Germany). A paired t-test was performed with GraphPad Prism6, and changes in the width of the amplitude range were compared between the two types of gating images. Results The following results were obtained from all patients when the respiratory gating method was applied: $SUV_{max}=9.43{\pm}3.93$, $SUV_{mean}=1.77{\pm}0.89$, and $Tumor\;Volume=4.17{\pm}2.41$ for the non-gating images, $SUV_{max}=10.08{\pm}4.07$, $SUV_{mean}=1.75{\pm}0.81$, and $Tumor\;Volume=3.56{\pm}2.11$ for the gating images, and $SUV_{max}=10.86{\pm}4.36$, $SUV_{mean}=1.77{\pm}0.85$, $Tumor\;Volume=3.36{\pm}1.98$ for the additional lung gating images. No statistically significant difference in the values of $SUV_{mean}$ was found between the non-gating and gating images, and between the gating and lung gating images (P>0.05). A significant difference in the values of $SUV_{max}$ and Tumor Volume were found between the aforementioned groups (P<0.05). The width of the amplitude range was smaller for lung gating images than gating images for 12 from 20 patients (3 patients with stable signals, 9 patients with unstable signals). Conclusion In PET/CT scanning using the respiratory gating method in the flow mode, any lesion movements caused by respiration were adjusted; therefore, more accurate measurements of $SUV_{max}$, and Tumor Volume could be obtained from the gating images than the non-gating images in this study. In addition, the width of the amplitude range decreased according to the stability of respiration to a more significant degree in the additional lung gating images than the gating images. We found that gating images provide information that is more useful for diagnosis than the one provided by non-gating images. For patients with irregular signals, it may be helpful to perform localized scanning additionally if time allows.

• Journal of Intelligence and Information Systems
• /
• v.25 no.2
• /
• pp.25-38
• /
• 2019

Selecting high-quality information that meets the interests and needs of users among the overflowing contents is becoming more important as the generation continues. In the flood of information, efforts to reflect the intention of the user in the search result better are being tried, rather than recognizing the information request as a simple string. Also, large IT companies such as Google and Microsoft focus on developing knowledge-based technologies including search engines which provide users with satisfaction and convenience. Especially, the finance is one of the fields expected to have the usefulness and potential of text data analysis because it's constantly generating new information, and the earlier the information is, the more valuable it is. Automatic knowledge extraction can be effective in areas where information flow is vast, such as financial sector, and new information continues to emerge. However, there are several practical difficulties faced by automatic knowledge extraction. First, there are difficulties in making corpus from different fields with same algorithm, and it is difficult to extract good quality triple. Second, it becomes more difficult to produce labeled text data by people if the extent and scope of knowledge increases and patterns are constantly updated. Third, performance evaluation is difficult due to the characteristics of unsupervised learning. Finally, problem definition for automatic knowledge extraction is not easy because of ambiguous conceptual characteristics of knowledge. So, in order to overcome limits described above and improve the semantic performance of stock-related information searching, this study attempts to extract the knowledge entity by using neural tensor network and evaluate the performance of them. Different from other references, the purpose of this study is to extract knowledge entity which is related to individual stock items. Various but relatively simple data processing methods are applied in the presented model to solve the problems of previous researches and to enhance the effectiveness of the model. From these processes, this study has the following three significances. First, A practical and simple automatic knowledge extraction method that can be applied. Second, the possibility of performance evaluation is presented through simple problem definition. Finally, the expressiveness of the knowledge increased by generating input data on a sentence basis without complex morphological analysis. The results of the empirical analysis and objective performance evaluation method are also presented. The empirical study to confirm the usefulness of the presented model, experts' reports about individual 30 stocks which are top 30 items based on frequency of publication from May 30, 2017 to May 21, 2018 are used. the total number of reports are 5,600, and 3,074 reports, which accounts about 55% of the total, is designated as a training set, and other 45% of reports are designated as a testing set. Before constructing the model, all reports of a training set are classified by stocks, and their entities are extracted using named entity recognition tool which is the KKMA. for each stocks, top 100 entities based on appearance frequency are selected, and become vectorized using one-hot encoding. After that, by using neural tensor network, the same number of score functions as stocks are trained. Thus, if a new entity from a testing set appears, we can try to calculate the score by putting it into every single score function, and the stock of the function with the highest score is predicted as the related item with the entity. To evaluate presented models, we confirm prediction power and determining whether the score functions are well constructed by calculating hit ratio for all reports of testing set. As a result of the empirical study, the presented model shows 69.3% hit accuracy for testing set which consists of 2,526 reports. this hit ratio is meaningfully high despite of some constraints for conducting research. Looking at the prediction performance of the model for each stocks, only 3 stocks, which are LG ELECTRONICS, KiaMtr, and Mando, show extremely low performance than average. this result maybe due to the interference effect with other similar items and generation of new knowledge. In this paper, we propose a methodology to find out key entities or their combinations which are necessary to search related information in accordance with the user's investment intention. Graph data is generated by using only the named entity recognition tool and applied to the neural tensor network without learning corpus or word vectors for the field. From the empirical test, we confirm the effectiveness of the presented model as described above. However, there also exist some limits and things to complement. Representatively, the phenomenon that the model performance is especially bad for only some stocks shows the need for further researches. Finally, through the empirical study, we confirmed that the learning method presented in this study can be used for the purpose of matching the new text information semantically with the related stocks.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.28 no.2
• /
• pp.216-226
• /
• 1983

Genetic studies on the $F_2$ generation of a set of half diallel crosses involving six cowpea varieties were conducted. by the randomized block design with three replications to determine combining ability, gene action and the relationships between parents and their $F_2$ hybrids. The 12 agronomic characters namely, days to flowering, days from flowering to maturity, days to maturity, diameter of stem, length of internode, number of branches per plant, length of pod, number of pods per plant, number of grains per pod, number of grains per plant, 100 grain weight and grain weight per plot were observed, and the $F_2$ generation of this diallel set of crosses was analysed for each character according to the method by Jinks and Hayman. The results obtained are summarized as follows: 1. Vr-Wr graphical analyses; The following seven characters, days to flowering, number of branches per plant, length of pod, number of pods per plant, number of grains per plant, 100 grain weight and grain weight per plot appeared to be partially dominant, and over dominance was found for days from flowering to maturity, days to maturity, length of internode and number of grains per pod. But diameter of stem indicated partial dominance near complete dominance. 2. Estimates of genetic variance components; In the degree of dominance,. eight characters, that is, days to flowering, days from flowering to maturity, days to maturity, length of internode, number of pods per plant, number of grains per pod, number of grains per plant and grain weight per plot were expressed larger than 1. And the characters, days from flowering to maturity, number of branches per plant and number of grains per plant as the degree of mean dominance ($H_1$/D) were found to be negative value over other characters. On the other hand, apprent asymmetry of dominance-recessive allele ($H_2$ /$4H_1$) produced comparatively estimates with lower value on days from flowering to maturity, length of internode, number of branches per plant and number of grains per pod. 3. Analyses of combining ability; Mean square value of GCA(general combining ability) appeared to be more important than those of SCA (specific combining ability) for most characters, and among them, grain weight per plot showed the highest mean square value in GCA and SCA. 4. Effect of combining ability; Variety 178 was expressed as the highest GCA effects in five characters (days to flowering days to maturity, number of pods per plant, number of grains per plant and grain weight per plot). SCA effects were differed from parents, characters and crosses, but crosses between TVu 1857 $\times$ TVu 2885 and TVu 2702 $\times$ J78 were shown to be highly with SCA effects on yield.

• Journal of Intelligence and Information Systems
• /
• v.28 no.1
• /
• pp.107-129
• /
• 2022

It is important to preempt new technology because the technology competition is getting much tougher. Stakeholders conduct exploration activities continuously for new technology preoccupancy at the right time. Gartner's Hype Cycle has significant implications for stakeholders. The Hype Cycle is a expectation graph for new technologies which is combining the technology life cycle (S-curve) with the Hype Level. Stakeholders such as R&D investor, CTO(Chef of Technology Officer) and technical personnel are very interested in Gartner's Hype Cycle for new technologies. Because high expectation for new technologies can bring opportunities to maintain investment by securing the legitimacy of R&D investment. However, contrary to the high interest of the industry, the preceding researches faced with limitations aspect of empirical method and source data(news, academic papers, search traffic, patent etc.). In this study, we focused on two research questions. The first research question was 'Is there a difference in the characteristics of the network structure at each stage of the hype cycle?'. To confirm the first research question, the structural characteristics of each stage were confirmed through the component cohesion size. The second research question is 'Is there a pattern of diffusion at each stage of the hype cycle?'. This research question was to be solved through centralization index and network density. The centralization index is a concept of variance, and a higher centralization index means that a small number of nodes are centered in the network. Concentration of a small number of nodes means a star network structure. In the network structure, the star network structure is a centralized structure and shows better diffusion performance than a decentralized network (circle structure). Because the nodes which are the center of information transfer can judge useful information and deliver it to other nodes the fastest. So we confirmed the out-degree centralization index and in-degree centralization index for each stage. For this purpose, we confirmed the structural features of the community and the expectation diffusion patterns using Social Network Serice(SNS) data in 'Gartner Hype Cycle for Artificial Intelligence, 2021'. Twitter data for 30 technologies (excluding four technologies) listed in 'Gartner Hype Cycle for Artificial Intelligence, 2021' were analyzed. Analysis was performed using R program (4.1.1 ver) and Cyram Netminer. From October 31, 2021 to November 9, 2021, 6,766 tweets were searched through the Twitter API, and converting the relationship user's tweet(Source) and user's retweets (Target). As a result, 4,124 edgelists were analyzed. As a reult of the study, we confirmed the structural features and diffusion patterns through analyze the component cohesion size and degree centralization and density. Through this study, we confirmed that the groups of each stage increased number of components as time passed and the density decreased. Also 'Innovation Trigger' which is a group interested in new technologies as a early adopter in the innovation diffusion theory had high out-degree centralization index and the others had higher in-degree centralization index than out-degree. It can be inferred that 'Innovation Trigger' group has the biggest influence, and the diffusion will gradually slow down from the subsequent groups. In this study, network analysis was conducted using social network service data unlike methods of the precedent researches. This is significant in that it provided an idea to expand the method of analysis when analyzing Gartner's hype cycle in the future. In addition, the fact that the innovation diffusion theory was applied to the Gartner's hype cycle's stage in artificial intelligence can be evaluated positively because the Gartner hype cycle has been repeatedly discussed as a theoretical weakness. Also it is expected that this study will provide a new perspective on decision-making on technology investment to stakeholdes.

• Journal of radiological science and technology
• /
• v.31 no.1
• /
• pp.17-23
• /
• 2008

Purpose : Because there is a difference depending on the environment as for an inspection equipment the important part of bone density scan and the precision/accuracy of a tester, the management of quality must be made systematically. The equipment failure caused by overload effect due to the aged equipment and the increase of a patient was made frequently. Thus, the replacement of equipment and additional purchases of new bonedensity equipment caused a compatibility problem in tracking patients. This study wants to know whether the clinical changes of patient's bonedensity can be accurately and precisely reflected when used it compatiblly like the existing equipment after equipment replacement and expansion. Materials and methods : Two equipments of GE Lunar Prodigy Advance(P1 and P2) and the Phantom HOLOGIC Spine Road(HSP) were used to measure equipment precision. Each device scans 20 times so that precision data was acquired from the phantom(Group 1). The precision of a tester was measured by shooting twice the same patient, every 15 members from each of the target equipment in 120 women(average age 48.78, 20-60 years old)(Group 2). In addition, the measurement of the precision of a tester and the cross-calibration data were made by scanning 20 times in each of the equipment using HSP, based on the data obtained from the management of quality using phantom(ASP) every morning (Group 3). The same patient was shot only once in one equipment alternately to make the measurement of the precision of a tester and the cross-calibration data in 120 women(average age 48.78, 20-60 years old)(Group 4). Results : It is steady equipment according to daily Q.C Data with $0.996\;g/cm^2$, change value(%CV) 0.08. The mean${\pm}$SD and a %CV price are ALP in Group 1(P1 : $1.064{\pm}0.002\;g/cm^2$, $%CV=0.190\;g/cm^2$, P2 : $1.061{\pm}0.003\;g/cm^2$, %CV=0.192). The mean${\pm}$SD and a %CV price are P1 : $1.187{\pm}0.002\;g/cm^2$, $%CV=0.164\;g/cm^2$, P2 : $1.198{\pm}0.002\;g/cm^2$, %CV=0.163 in Group 2. The average error${\pm}$2SD and %CV are P1 - (spine: $0.001{\pm}0.03\;g/cm^2$, %CV=0.94, Femur: $0.001{\pm}0.019\;g/cm^2$, %CV=0.96), P2 - (spine: $0.002{\pm}0.018\;g/cm^2$, %CV=0.55, Femur: $0.001{\pm}0.013\;g/cm^2$, %CV=0.48) in Group 3. The average error${\pm}2SD$, %CV, and r value was spine : $0.006{\pm}0.024\;g/cm^2$, %CV=0.86, r=0.995, Femur: $0{\pm}0.014\;g/cm^2$, %CV=0.54, r=0.998 in Group 4. Conclusion: Both LUNAR ASP CV% and HOLOGIC Spine Phantom are included in the normal range of error of ${\pm}2%$ defined in ISCD. BMD measurement keeps a relatively constant value, so showing excellent repeatability. The Phantom has homogeneous characteristics, but it has limitations to reflect the clinical part including variations in patient's body weight or body fat. As a result, it is believed that quality control using Phantom will be useful to check mis-calibration of the equipment used. A value measured a patient two times with one equipment, and that of double-crossed two equipment are all included within 2SD Value in the Bland - Altman Graph compared results of Group 3 with Group 4. The r value of 0.99 or higher in Linear regression analysis(Regression Analysis) indicated high precision and correlation. Therefore, it revealed that two compatible equipment did not affect in tracking the patients. Regular testing equipment and capabilities of a tester, then appropriate calibration will have to be achieved in order to calculate confidential BMD.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.23 no.2
• /
• pp.1-24
• /
• 1978

To obtain basic information on the breeding of early maturing, short culm naked-barley varieties, the following 10 varieties, Ehime ＃ 1, Shikoku ＃42, Yamate hadaka, Eijo hadaka, Kagawa ＃ 1, Jangjubaeggwa, Baegdong, Cheongmaeg, Seto-hadaka and Mokpo ＃42 were used in diallel crosses in 1974. Heading date, culm length and grain yield per plant for the parents, $F_1's$ and $F_2's$ of the 10X10 partial diallel crosses were measured in 1976 for analysis of their combining ability, heritability and inheritance. The results obtained are summarized below; 1. Heritabilities in broad sense for heading date, culm length and grain yield per plant were 0.7831, 0.7599 and 0.6161, respectively. Narrow sense heritabilities for heading date were 0.3972 in $F_1$ and 0.7789 in $F_2$ and for culm length 0.6567 in $F_1$ and 0.6414 in $F_2.$ These values suggest that earliness and culm length could be successfully selected for in the early generations. Narrow sense heritability for grain yield was 0.3775 in $F_1$ and 0.4170 in $F_2.$ 2. GCA effects of the $F_1$ and $F_2$ generations for days to heading were high in the early direction for early-heading varieties, while for late-heading varieties the GCA effects were high in the late direction. Absolute values for GCA effects in $F_1$ were higher than in $F_2.$ SCA effects of the $F_1$ and $F_2$ generations were high in the early-heading direction for Shikoku ＃ 42 x Mokpo ＃ 42, Ehime ＃ 1 x Yamate hadaka, Shikoku ＃ 42 x Yamate hadaka and Shikoku ＃42 x Eijo hadaka. 3. The GCA effects for culm length in the $F_1$ and $F_2$ generations for tall varieties were high in the tall direction while short varieties were high in the short direction. Absolute values for the GCA effects in $F_1$ were higher than in $F_2.$ SCA effects were high in the short direction for the combinations of Mokpo ＃ 42 with Ehime ＃ 1, Yamate had aka and Eijo hadaka. 4. The GCA effects for grain yields per plant in the $F_1$ and $F_2$ generations for varieties with high yields per plant were high in the high yielding direction, while varieties with low yields per plant were high in the low yielding direction. Absolute values of the $F_1$ GCA effects were higher than the $F_2$ effects. The combinations with high SCA effects were Mokpo ＃ 42 x Shikoku ＃ 42, Mokpo ＃ 42 x Seto hadaka and Mokpo ＃ 42 x Cheongmaeg. 5. Mean heading dates of the $F_1$ and $F_2$ generations were earlier than those of mean mid-parent. Mean heading date of the $F_1$ generation was earlier than the $F_2$ generation. Crosses involving early-heading varieties showed a greater $F_1, $ mid-parent difference than crosses involving late-heading varieties. 6. Heading date was controlled by a partial dominance effect. Nine varieties excluding Mokpo ＃ 42 showed allelic gene action. Ehime ＃ 1, Shikoku ＃ 42, Kagawa ＃ 1 and Mokpo ＃ 42 were recessive to the other tested varieties. 7. The $F_2$ segregations of the 45 crosses for days to heading showed that 33 cosses were of such complexity that they could not be explained by simple genetic inheritance. One cross showed a 3 : 1 ratio where earliness was dominant. Another cross showed a 3 : 1 ratio where lateness was dominant. Four other crosses showed a 9 : 7 ratio for earliness while six crosses showed a 9 : 7 ratio for lateness. 8. Many transgressive segregants for earliness were found in the following crosses; Eijo hadaka x Baegdong, Ehime ＃ 1 x Seto hadaka, Yamate had aka x Kagawa ＃ 1, Kagawa ＃ 1 x Sato hadaka, Shikoku ＃ 42 x Kagawa ＃ 1, Ehime ＃ 1 x Kagawa ＃ 1, Ehime ＃ 1 x Shikoku ＃ 42, Ehime ＃ 1 x Eijo hadaka. 9. Mean culm length of the F, and F. generations were usually taller than the mid-parent where tall parent were used. These trends were high in the short varieties, but low in the tall varieties. 10. Culm length was controlled by partial dominace which was gonverned by allelic gene(s). Culm length showed a high degree of control by additive genes. Mokpo ＃ 42 was recessive while Baegdong was dominant. 11. The F_2 frequency for culm length was in large part normally distributed around the midparent value. However, some combinations showed transgressive segregation for either tall or short culm length. From combinations between medium tall varieties, Ehime ＃ 1, Shikoku ＃ 42, Eijo hadaka and Seto hadaka, many short segregants could be found. 12. Mean grain yields per plant of the F_1 and F_2 generations were 6% and 5% higher than those of mid-parents, respectively. The varieties with high yields per plant showed a low rate of yield increase in their F_1's and F_2's while the varieties with low yields per plant showed a high rate of yield increase in their F_1's and F_1's. 13. Grain yields per plant showed over-dominnee effects, governed by non-allelic genes. Mokpo ＃ 42 showed recessive genetic control of grain yield per plant. It remains difficult to clarify the inheritance of grain yields per plant from these data.

• Journal of Intelligence and Information Systems
• /
• v.19 no.2
• /
• pp.1-20
• /
• 2013

Nowadays, social network is a huge communication platform for providing people to connect with one another and to bring users together to share common interests, experiences, and their daily activities. Users spend hours per day in maintaining personal information and interacting with other people via posting, commenting, messaging, games, social events, and applications. Due to the growth of user's distributed information in social network, there is a great potential to utilize the social data to enhance the quality of recommender system. There are some researches focusing on social network analysis that investigate how social network can be used in recommendation domain. Among these researches, we are interested in taking advantages of the interaction between a user and others in social network that can be determined and known as social relationship. Furthermore, mostly user's decisions before purchasing some products depend on suggestion of people who have either the same preferences or closer relationship. For this reason, we believe that user's relationship in social network can provide an effective way to increase the quality in prediction user's interests of recommender system. Therefore, social relationship between users encountered from social network is a common factor to improve the way of predicting user's preferences in the conventional approach. Recommender system is dramatically increasing in popularity and currently being used by many e-commerce sites such as Amazon.com, Last.fm, eBay.com, etc. Collaborative filtering (CF) method is one of the essential and powerful techniques in recommender system for suggesting the appropriate items to user by learning user's preferences. CF method focuses on user data and generates automatic prediction about user's interests by gathering information from users who share similar background and preferences. Specifically, the intension of CF method is to find users who have similar preferences and to suggest target user items that were mostly preferred by those nearest neighbor users. There are two basic units that need to be considered by CF method, the user and the item. Each user needs to provide his rating value on items i.e. movies, products, books, etc to indicate their interests on those items. In addition, CF uses the user-rating matrix to find a group of users who have similar rating with target user. Then, it predicts unknown rating value for items that target user has not rated. Currently, CF has been successfully implemented in both information filtering and e-commerce applications. However, it remains some important challenges such as cold start, data sparsity, and scalability reflected on quality and accuracy of prediction. In order to overcome these challenges, many researchers have proposed various kinds of CF method such as hybrid CF, trust-based CF, social network-based CF, etc. In the purpose of improving the recommendation performance and prediction accuracy of standard CF, in this paper we propose a method which integrates traditional CF technique with social relationship between users discovered from user's behavior in social network i.e. Facebook. We identify user's relationship from behavior of user such as posts and comments interacted with friends in Facebook. We believe that social relationship implicitly inferred from user's behavior can be likely applied to compensate the limitation of conventional approach. Therefore, we extract posts and comments of each user by using Facebook Graph API and calculate feature score among each term to obtain feature vector for computing similarity of user. Then, we combine the result with similarity value computed using traditional CF technique. Finally, our system provides a list of recommended items according to neighbor users who have the biggest total similarity value to the target user. In order to verify and evaluate our proposed method we have performed an experiment on data collected from our Movies Rating System. Prediction accuracy evaluation is conducted to demonstrate how much our algorithm gives the correctness of recommendation to user in terms of MAE. Then, the evaluation of performance is made to show the effectiveness of our method in terms of precision, recall, and F1-measure. Evaluation on coverage is also included in our experiment to see the ability of generating recommendation. The experimental results show that our proposed method outperform and more accurate in suggesting items to users with better performance. The effectiveness of user's behavior in social network particularly shows the significant improvement by up to 6% on recommendation accuracy. Moreover, experiment of recommendation performance shows that incorporating social relationship observed from user's behavior into CF is beneficial and useful to generate recommendation with 7% improvement of performance compared with benchmark methods. Finally, we confirm that interaction between users in social network is able to enhance the accuracy and give better recommendation in conventional approach.

• Journal of Intelligence and Information Systems
• /
• v.19 no.1
• /
• pp.57-77
• /
• 2013

Social media has become one of the most popular media in web and mobile application. In 2011, social networks and blogs are still the top destination of online users, according to a study from Nielsen Company. In their studies, nearly 4 in 5active users visit social network and blog. Social Networks and Blogs sites rule Americans' Internet time, accounting to 23 percent of time spent online. Facebook is the main social network that the U.S internet users spend time more than the other social network services such as Yahoo, Google, AOL Media Network, Twitter, Linked In and so on. In recent trend, most of the companies promote their products in the Facebook by creating the "Facebook Page" that refers to specific product. The "Like" option allows user to subscribed and received updates their interested on from the page. The film makers which produce a lot of films around the world also take part to market and promote their films by exploiting the advantages of using the "Facebook Page". In addition, a great number of streaming service providers allows users to subscribe their service to watch and enjoy movies and TV program. They can instantly watch movies and TV program over the internet to PCs, Macs and TVs. Netflix alone as the world's leading subscription service have more than 30 million streaming members in the United States, Latin America, the United Kingdom and the Nordics. As the matter of facts, a million of movies and TV program with different of genres are offered to the subscriber. In contrast, users need spend a lot time to find the right movies which are related to their interest genre. Recent years there are many researchers who have been propose a method to improve prediction the rating or preference that would give the most related items such as books, music or movies to the garget user or the group of users that have the same interest in the particular items. One of the most popular methods to build recommendation system is traditional Collaborative Filtering (CF). The method compute the similarity of the target user and other users, which then are cluster in the same interest on items according which items that users have been rated. The method then predicts other items from the same group of users to recommend to a group of users. Moreover, There are many items that need to study for suggesting to users such as books, music, movies, news, videos and so on. However, in this paper we only focus on movie as item to recommend to users. In addition, there are many challenges for CF task. Firstly, the "sparsity problem"; it occurs when user information preference is not enough. The recommendation accuracies result is lower compared to the neighbor who composed with a large amount of ratings. The second problem is "cold-start problem"; it occurs whenever new users or items are added into the system, which each has norating or a few rating. For instance, no personalized predictions can be made for a new user without any ratings on the record. In this research we propose a clustering method according to the users' genre interest extracted from social network service (SNS) and user's movies rating information system to solve the "cold-start problem." Our proposed method will clusters the target user together with the other users by combining the user genre interest and the rating information. It is important to realize a huge amount of interesting and useful user's information from Facebook Graph, we can extract information from the "Facebook Page" which "Like" by them. Moreover, we use the Internet Movie Database(IMDb) as the main dataset. The IMDbis online databases that consist of a large amount of information related to movies, TV programs and including actors. This dataset not only used to provide movie information in our Movie Rating Systems, but also as resources to provide movie genre information which extracted from the "Facebook Page". Formerly, the user must login with their Facebook account to login to the Movie Rating System, at the same time our system will collect the genre interest from the "Facebook Page". We conduct many experiments with other methods to see how our method performs and we also compare to the other methods. First, we compared our proposed method in the case of the normal recommendation to see how our system improves the recommendation result. Then we experiment method in case of cold-start problem. Our experiment show that our method is outperform than the other methods. In these two cases of our experimentation, we see that our proposed method produces better result in case both cases.