• Journal of Intelligence and Information Systems
• /
• v.29 no.1
• /
• pp.249-263
• /
• 2023

Collaborative Filtering, a representative recommendation system methodology, consists of two approaches: neighbor methods and latent factor models. Among these, the latent factor model using matrix factorization decomposes the user-item interaction matrix into two lower-dimensional rectangular matrices, predicting the item's rating through the product of these matrices. Due to the factor vectors inferred from rating patterns capturing user and item characteristics, this method is superior in scalability, accuracy, and flexibility compared to neighbor-based methods. However, it has a fundamental drawback: the need to reflect the diversity of preferences of different individuals for items with no ratings. This limitation leads to repetitive and inaccurate recommendations. The Adaptive Deep Latent Factor Model (ADLFM) was developed to address this issue. This model adaptively learns the preferences for each item by using the item description, which provides a detailed summary and explanation of the item. ADLFM takes in item description as input, calculates latent vectors of the user and item, and presents a method that can reflect personal diversity using an attention score. However, due to the requirement of a dataset that includes item descriptions, the domain that can apply ADLFM is limited, resulting in generalization limitations. This study proposes a Generalized Adaptive Deep Latent Factor Recommendation Model, G-ADLFRM, to improve the limitations of ADLFM. Firstly, we use item ID, commonly used in recommendation systems, as input instead of the item description. Additionally, we apply improved deep learning model structures such as Self-Attention, Multi-head Attention, and Multi-Conv1D. We conducted experiments on various datasets with input and model structure changes. The results showed that when only the input was changed, MAE increased slightly compared to ADLFM due to accompanying information loss, resulting in decreased recommendation performance. However, the average learning speed per epoch significantly improved as the amount of information to be processed decreased. When both the input and the model structure were changed, the best-performing Multi-Conv1d structure showed similar performance to ADLFM, sufficiently counteracting the information loss caused by the input change. We conclude that G-ADLFRM is a new, lightweight, and generalizable model that maintains the performance of the existing ADLFM while enabling fast learning and inference.

• Proceedings of the Korean Society for Technology of Plasticity Conference
• /
• 2008.10a
• /
• pp.367-369
• /
• 2008

A theoretical model of equivalent drawbead for sheet metal forming analysis is experimentally verified in this paper. After the theoretical drawbead models improved a material description for the accurate calculation of drawbead forces are briefly introduced, they are verified by showing the good agreement of their drawbead forces with experimental measurements. Furthermore, the excellence of theoretical models is demonstrated by the comparison with those of commercial codes.

• Proceedings of the Korea Institute of Fire Science and Engineering Conference
• /
• 1996.11a
• /
• pp.15-18
• /
• 1996

A general description of sprinkler activation time in compartment-fire-generated smoke layers is made. For calculation of the time hot layer temperature is obtained from two-layer zonal model and time constant of sprinkler is measured. Upper-layer thickness at the instant of sprinkler activation is also presented with changes of opening area. The outputs of the present study provide inputs for the interaction modeling of sprinkler spray and compartment fire environment, which simulates fire suppression phenomena.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.14 no.1
• /
• pp.1-4
• /
• 1977

In the description and design of certain kinds of sequential machines, we often meet inadequacies of traditional models we adopt to use. To Lessen these inconveniencies, this paper aims to define polymachine as a new basic model of sequential machine. Polymachine, defind on the level of state model, is composed of an index which is and independent componet and a polymer which, as a dependent component, consists of many element machines. In the meantime, some basic characters of the machine are illustrated and discussed in detail, in order to present the usefulness of polymachine.

• Proceedings of the IEEK Conference
• /
• 1998.10a
• /
• pp.755-758
• /
• 1998

As the performance of microprocessor improves, the design complexity grows exponentially. Therefor, it is very important to make the bug-free model as early as possible in a design life-cycle. This paper describes the simulation-based functional verification methodology for the RTL level description model. It is performed by multi-stage verification methods using extensive hand-generated self-checking tests supplemented with random tests. This approach is opplied to the functional verification of the GPU processor of Raptor and various bugs are detected.

• Proceedings of the Korean Society of Agricultural Engineers Conference
• /
• 2002.10a
• /
• pp.329-332
• /
• 2002

In the present study, in order to apply a cyclic viscoelastic-viscoplastic constitutive model to multi-layered ground conditions during large earthquake, the numerical simulations of the 1995 Hyogoken Nanbu Earthquake at Port Island, Kobe, Japan, were performed by the seismic respons analysis. From the seismic response analysis, it was verified that a cyclic viscoelastic-viscoplastic constitutive model can give a good description of the damping characteristics of clay accurately during large event which induces plastic deformation in large strain range.

• Journal of applied mathematics & informatics
• /
• v.15 no.1_2
• /
• pp.415-423
• /
• 2004

In allelic model X = ($x_1,\;x_2,...x_{d}$), $M_f(t)$= f(p(t)) - ${{\int}^{t}}_0$Lf(p(t))ds is a P-martingale for diffusion operator L under the certain conditions. In this note, we can show uniqueness of martingale problem associated with mean vector and obtain a complete description of ergodic property by using of the semigroup method.

• Journal of the Korean Geophysical Society
• /
• v.8 no.3
• /
• pp.153-157
• /
• 2005

A discussion of a method has been used with success in terrain modelling to estimate the height at any point on the land surface from irregularly distributed samples. The special requirements of terrain modelling are discussed as well as a detailed description of the algorithm and an example of its application.

• Proceedings of the Korean Society of Fisheries Technology Conference
• /
• 2000.10a
• /
• pp.167-168
• /
• 2000

A North Pacific Ocean Model has been developed with the Global Ocean Model of the Meteorological Research Institute of Japan which solves the primitive equations with Boussinesq, rigid-lid, and hydrostatic assumptions. The objective of the study is to improve the description of the variability on the East Sea and northwestern Pacific Ocean. (omitted)

• Proceedings of the Korean Society of Coastal and Ocean Engineers Conference
• /
• 1991.07a
• /
• pp.6-10
• /
• 1991

The bottom friction model of combined wave-current flow developed by Bijker, Yoo and O'Conner (hereafter it is called BYO model) is fully based on the Prandtl's mixing length theory. Although the Prandtl's theory is dependant on some rationalism, it is widely recognized that the theory is generally acceptable for the description of any turbulent flow.(omitted)