• Journal of Intelligence and Information Systems
• /
• v.21 no.1
• /
• pp.119-142
• /
• 2015

With the explosive growth in the volume of information, Internet users are experiencing considerable difficulties in obtaining necessary information online. Against this backdrop, ever-greater importance is being placed on a recommender system that provides information catered to user preferences and tastes in an attempt to address issues associated with information overload. To this end, a number of techniques have been proposed, including content-based filtering (CBF), demographic filtering (DF) and collaborative filtering (CF). Among them, CBF and DF require external information and thus cannot be applied to a variety of domains. CF, on the other hand, is widely used since it is relatively free from the domain constraint. The CF technique is broadly classified into memory-based CF, model-based CF and hybrid CF. Model-based CF addresses the drawbacks of CF by considering the Bayesian model, clustering model or dependency network model. This filtering technique not only improves the sparsity and scalability issues but also boosts predictive performance. However, it involves expensive model-building and results in a tradeoff between performance and scalability. Such tradeoff is attributed to reduced coverage, which is a type of sparsity issues. In addition, expensive model-building may lead to performance instability since changes in the domain environment cannot be immediately incorporated into the model due to high costs involved. Cumulative changes in the domain environment that have failed to be reflected eventually undermine system performance. This study incorporates the Markov model of transition probabilities and the concept of fuzzy clustering with CBCF to propose predictive clustering-based CF (PCCF) that solves the issues of reduced coverage and of unstable performance. The method improves performance instability by tracking the changes in user preferences and bridging the gap between the static model and dynamic users. Furthermore, the issue of reduced coverage also improves by expanding the coverage based on transition probabilities and clustering probabilities. The proposed method consists of four processes. First, user preferences are normalized in preference clustering. Second, changes in user preferences are detected from review score entries during preference transition detection. Third, user propensities are normalized using patterns of changes (propensities) in user preferences in propensity clustering. Lastly, the preference prediction model is developed to predict user preferences for items during preference prediction. The proposed method has been validated by testing the robustness of performance instability and scalability-performance tradeoff. The initial test compared and analyzed the performance of individual recommender systems each enabled by IBCF, CBCF, ICFEC and PCCF under an environment where data sparsity had been minimized. The following test adjusted the optimal number of clusters in CBCF, ICFEC and PCCF for a comparative analysis of subsequent changes in the system performance. The test results revealed that the suggested method produced insignificant improvement in performance in comparison with the existing techniques. In addition, it failed to achieve significant improvement in the standard deviation that indicates the degree of data fluctuation. Notwithstanding, it resulted in marked improvement over the existing techniques in terms of range that indicates the level of performance fluctuation. The level of performance fluctuation before and after the model generation improved by 51.31% in the initial test. Then in the following test, there has been 36.05% improvement in the level of performance fluctuation driven by the changes in the number of clusters. This signifies that the proposed method, despite the slight performance improvement, clearly offers better performance stability compared to the existing techniques. Further research on this study will be directed toward enhancing the recommendation performance that failed to demonstrate significant improvement over the existing techniques. The future research will consider the introduction of a high-dimensional parameter-free clustering algorithm or deep learning-based model in order to improve performance in recommendations.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.4
• /
• pp.101-109
• /
• 2014

It is an essential element for the establishment of image processing related systems to find the exact contour from the image of an arbitrary object. In particular, if a vision system is established to inspect the products in the automated production process, it is very important to detect the contours for standardized shapes such lines and curves. In this paper, we propose a prototype adaptive dynamic contour model, p-Snake with improved contour extraction algorithms by adding the prototype energy. The proposed method is to find the initial contour by applying the existing Snake algorithm after Sobel operation is performed for prototype analysis. Next, the final contour of the object is detected by analyzing prototypes such as lines and circles, defining prototype energy and using it as an additional energy item in the existing Snake function on the basis of information on initial contour. We performed experiments on 340 images obtained by using an environment that duplicated the background of an industrial site. It was found that even if objects are not clearly distinguished from the background due to noise and lighting or the edges being insufficiently visible in the images, the contour can be extracted. In addition, in the case of similarity which is the measure representing how much it matches the prototype, the prototype similarity of contour extracted from the proposed p-ACM is superior to that of ACM by 9.85%.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.44 no.4 s.316
• /
• pp.26-34
• /
• 2007

Intelligent projection system indicates a system that displays desired images on an arbitrary screen in an arbitrary environment using projector without noticeable image distortion. In recent years, projectors have become widespread and ubiquitous due to their increasing capabilities and declining cost. Moreover, the size of projectors is getting smaller and handhold projectors are emerging. Thanks to these advances, the demand for intelligent projection system has been significantly increased and the demand has led to remarkable progress of the related techniques or technologies to intelligent projection system However, there are still some environments (or conditions, mainly dynamic ones) that intelligent projections systems cannot handle and they have limited the application area of intelligent projection systems. This paper exemplifies such environments (e.g. specular screen, dynamic screen) and propose effective solutions (i.e. multiple overlapping projectors, complementary pattern embedding) for thor And the usefulness of the solutions is verified through experimental results and user evaluation. Notice that the environments are considered not simultaneously but independently because it is impossible to consider them simultaneously by simply combining the solutions for each. Acually, a totally different solution would be necessary to consider them simultaneously. Therefore, we expect that the proposed methods would largely extend the application area of intelligent projection systems except for severely arbitrary environment.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.26 no.4
• /
• pp.704-713
• /
• 1997

This study was conducted to describe qualitatively the protein metabolism of pigs during growth depending on the feed protein quality and to describe quantitatively amino acids requirements, using a simulation model. The used model has a non-linear structure. In the used model, the protein utilization system of a pig, which is in the non-steady-state, is described with 15 flux equations and 11 differential equations and is composed with two compartments. Protein deposition(g/day) of pigs on the 30th, 60th, 90th, and 120th day of feeding duration with three-quality protein, beginning with body weight 20kg, were calculated according to the empirical model, PAF(the product of amino acid functions) of Menke, and was used as object function for the simulation. The mean of relative difference between the simulated protein deposition and PAF calculated values, lied in a range of 8.8%. The simulated protein deposition showed different behavior according to feed protein quality. In the high-quality protein, it showed paraboloidal form with extending growth simulation up to 150eh day. So the maximum of protein deposition was acquired on the 105th day of simulate growth time and then it decreased fast. In the low-quality protein, this form of protein deposition in the course of simulated growth did not appear until 150th day. The simulated protein mass also showed a difference in accordance with feed protein quality. The difference was small on the 30th day of simulated growth, but with duration of the simulated growth it was larger. On the 150th day the simulated protein deposition of high quality protein was 1.5 times higher as compared to the low-quality protein. The simulated protein synthesis and break-down rates(g/day) in the whole body showed a parallel behavior in the course of growth, according to feed protein quality. It was found that the improvement of feed protein quality increased protein deposition in the whole body through a increase of both protein synthesis and breakdown during growth. Also protein deposition efficiency, which was calculated from simulated protein deposition and protein synthesis, showed a difference in dependence on the protein qualify of feed protein. The protein deposition efficiency was higher in pigs fed with high quality protein, especially at the simulation time 30th day. But this phenomena disappeared with growth, so on the 150th day of growth, the protein deposition of the high feed protein quality was lowest among the three different quality of feed protein. The simulated total requirement of the 10 essential amino acids for the growth of pigs was 28.1(g/100g protein), similar to NRC. The requirement of lysine was 4.2(g/100g protein).