Constraint graph-based frequent pattern updating from temporal databases

There have been many kinds of association rule mining (ARM) algorithms, e.g., Apriori and FP-tree, to discover meaningful frequent patterns from a large dataset. Particularly, it is more difficult for such ARM algorithms to be applied for temporal databases which are continuously changing over time. Such algorithms are generally based on repeating time-consuming tasks, e.g., scanning databases. To deal with this problem, in this paper, we propose a constraint graph-based method for maintaining frequent patterns (FP) discovered from the temporal databases. Particularly, the constraint graph, which is represented as a set of constraint between two items, can be established by temporal persistency of the patterns. It means that some patterns can be used to build the constraint graph, when the patterns have been shown in a set of the FP. Two types of constraints can be generated by users and adaptation. Based on our scheme, we find that a large number of dataset has been efficiently reduced during mining process and the gathering information while updating.     

Uniformly subsampled ensemble (USE) for churn management: Theory and implementation

The present paper explores the possible application of a new ensemble model. The model, which is based on multiple SVM classifiers, is employed to address churner identification problems in the mobile telecommunication industry, a sector in which the role of customer retention program becomes increasingly important due to its very competitive business environment. In particular, the current study introduces a uniformly subsampled ensemble (USE) model of SVM classifiers, not only to reduce the computational complexity of large-scale data, but also to boost the reliability and accuracy of calibrated models on data sets with highly skewed class distributions. According to our experiments, the performance of the USE SVM model is superior compared to all single and ensemble models. It is more scalable than well-known ensemble models as well.     

Parametric study on design of composite–foam–resin concrete sandwich structures for precision machine tool structures

In this paper, sandwich structures for micro-EDM machines are optimized by using parametric study varying composite geometries and parameters like stacking sequence, thickness and rib geometry. The structures are composed of fibre reinforced composites for skin material and resin concrete and PVC foam (Closed cell, Divinycell) for core materials. Column structure was designed by a beam with cruciform rib and performance indices such as static bending stiffness (EI) and specific bending stiffness (EI/ρ) for dynamic stability are examined by controlling the thickness and stacking sequence of composites. For the machine tool bed, which usually has a plate shape, was designed to have high stiffness in two directions at the same time controlling stacking sequence and rib geometry; that is, rib thickness and number of ribs. The sensitivity of design parameters like rib thickness and composite skin thickness was examined and the optimal condition for high stiffness structure was suggested. Finite element analysis was also performed to verify the static and dynamic robustness of the machine structure. L-shaped joint for combining bed and column of the micro-EDM machine was proposed and fabricated using adhesive bonding. The dynamic performance such as damping characteristics was investigated by vibration tests. From the results optimal configuration and materials for high precision micro-EDM machines are proposed.     

Material properties and growth control of undoped and Sn-doped In2O3 thin films prepared by using ion beam technologies

Undoped (IO) and Sn-doped In₂O₃ (ITO) films have been deposited on glass and polymer substrates by an advanced ion beam technologies including ion-assisted deposition (IAD), hybrid ion beam, ion beam sputter deposition (IBSD), and ion-assisted reaction (IAR). Physical and chemical properties of the oxide films and adhesion between films and substrates were improved significantly by these technologies. By using the IAD method, non-stoichiometry and microstructure of the films were controlled by changing assisted oxygen ion energy and arrival ratio of assisted oxygen ion to evaporated atoms. Relationships between structural and electrical properties in ITO films on glass substrates were intensively investigated by using the IBSD method with changing ion energy, reactive gas environment, and substrate temperature. Smooth-surface ITO films (R_rms ≤ 1 nm and R_p-v ≤ 10 nm) for organic light-emitting diodes were developed with a combination of deposition conditions with controlling microstructure of a seed layer on glass. IAR surface treatment enormously enhanced the adhesion of oxide films to polymer substrate. The different dependence of IO and ITO films' properties on the experimental parameters, such as ion energy and oxygen gas environment, will be intensively discussed.     

The preparation and characterization of porous carbons for hydrogen storage

Porous carbon materials with a cylindrical pore structure were prepared using ordered mesoporous silica as a removable template. To investigate the effect of the structural and textural properties of the products on hydrogen adsorption capacity, different carbon precursors and synthetic methods were used in their preparation. All of the carbon materials prepared showed a well-defined pore structure with a high surface area irrespective of the carbon precursor used in the preparation. Hydrogen adsorption tests indicated that the capacity of the materials for hydrogen adsorption was highly dependent on total surface area and the pore structure. Based on the N₂ sorption results, the total surface area was directly correlated with the hydrogen adsorption capacity.     

Tunable multiwavelength fiber laser employing a comb filter based on a polarization-diversity loop configuration

A tunable multiwavelength fiber laser based on a semiconductor optical amplifier (SOA) is newly proposed. The proposed fiber laser employs a fiber comb filter based on a polarization-diversity loop configuration (PDLC) for the multiwavelength selection. The spectral characteristic of the comb filter is graphically explained by tracing the state of polarization (SOP) of the light propagating through the filter on the Poincare sphere. Particularly, the method of adjusting the filter characteristics is established with the help of the Poincare-sphere representation. The experimental results show that 18 laser lines oscillate with the signal to amplified spontaneous emission noise ratio over 25 dB. Particularly, all of the laser lines with a fixed channel spacing of /spl sim/ 0.8 nm can be continuously shifted by adjusting only a wave plate contained within the filter.     

Low-Complexity Symbol Detector for MIMO-OFDM-Based Wireless LANs

In this brief, a low-complexity hardware architecture for multiple-input multiple-output (MIMO) orthogonal frequency-division multiplexing (OFDM) symbol detectors with two transmit and two receive antennas is proposed. The detectors support two MIMO-OFDM schemes of space-frequency block coded OFDM and space-division multiplexing OFDM in order to achieve higher performance and throughput. However, symbol detection processes for these two schemes have high computational complexity, which is a burden to hardware implementation of MIMO-OFDM symbol detectors. In order to reduce complexity, the proposed symbol detector is designed with shared architecture, where similar functional blocks are merged and share the hardware resources, and results in the reduction of logic gates by 34% over a conventional architecture employing two individual detectors     

Phase evolution and dielectric properties of MgTi2O5 ceramic sintered with lithium borosilicate glass

Phase evolution, densification, and dielectric properties of MgTi₂O₅ dielectric ceramic, sintered with lithium borosilicate (LBS) glass, were studied. Reaction between LBS glass and MgTi₂O₅ was significant in forming secondary phases such as TiO₂ and (Mg,Ti)₂(BO₃)O. The glass addition was not necessarily deleterious to the dielectric properties due to the formation of TiO₂: permittivity increased and temperature coefficient of resonance frequency could be tuned to zero with the addition of LBS glass, although the inevitable glass-induced decrease of quality factor was not retarded by the formation of TiO₂. The sintered specimen with 10 wt% LBS fired at 950 °C for 2 h showed permittivity of 19.3, quality factor of 6800 GHz, and τ_f of −16 ppm/°C.     

Optimal decoder for block-transform based video coders

In this paper, we introduce a new decoding algorithm for DCT-based video encoders, such as Motion JPEG (M-JPEG), H26x, or MPEG. This algorithm considers not only the compression artifacts but also the ones due to transmission, acquisition or storage of the video. The novelty of our approach is to jointly tackle these two problems, using a variational approach. The resulting decoder is object-based, allowing independent and adaptive processing of objects and backgrounds, and considers available information provided by the bitstream, such as quantization steps, and motion vectors. Several experiments demonstrate the efficiency of the proposed method. Objective and subjective quality assessment methods are used to evaluate the improvement upon standard algorithms, such as the deblocking and deringing filters included in MPEG-4 postprocessing.     

Expert system for scheduling in an airline gate allocation

Scheduling is an important technique encompassing a wide application area. Because of the complex interrelations among the resources, knowledge, and various other constraints, scheduling has many difficulties. Artificial Intelligence technology has been applied to solve the scheduling problem. As AI techniques are efficient in representing knowledge and dealing with heuristics, it is an adequate approach to model and to solve scheduling problems. We have implemented the ramp scheduling system, called RACES (Ramp Activity Coordination Expert System), to solve complex and dynamic aircraft parking problems. RACES was developed from the domain knowledge and experience which were acquired from the domain experts. Domain knowledge and experience are important factors in controlling the scheduling procedure. RACES divides the problem into sub-problems and experimental heuristics in the knowledge acquisition process. The system independently processes scheduling for the divided sub-problems and shares variables and domains. During the scheduling, the system selects or confines the search space with domain filtering techniques by exploiting the characteristics of various constraints and knowledge. RACES produces a user-driven near-optimal solution by means of a trade-off scheduling method using heuristics between the size of aircraft and the best-fit time. For 400 daily flights, RACES made parking schedules for aircraft in about 20 s compared with 4–5 h by human experts.