Jongjin Jung

Brokerage between buyer and seller agents using constraint satisfaction problem models

Abstract We have proposed a two-layered multi-agent framework for brokerage between buyers and sellers. The brokerage is processed in two layers for efficient linking between buyers and sellers: the competition layer and the constraint satisfaction layer. In the competition layer, we match the constituents of brokerage process to agents and use a functional relationship of a multi-agent framework for the communication among them. The seller agents, as requested by the buyer agents, are selected through the competition in the competition layer. In the constraint satisfaction layer, we model the relationship between buyer agents and seller agents in Constraint Satisfaction Problems (CSP). The CSP solver finds an optimal solution by choosing the best brokerage to satisfy various preferential requirements for users. We have implemented a prototype system for dealing in real estate on the Internet by applying the proposed framework. Finally, we have modeled the brokerage system using CSP and have shown the experimentation about the satisfaction level of users regarding their preferences.

Expert system for scheduling in an airline gate allocation

Abstract 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.

Effects of substrate conductivity on cell morphogenesis and proliferation using tailored, atomic layer deposition-grown ZnO thin films

We demonstrate that ZnO films grown by atomic layer deposition (ALD) can be employed as a substrate to explore the effects of electrical conductivity on cell adhesion, proliferation, and morphogenesis. ZnO substrates with precisely tunable electrical conductivity were fabricated on glass substrates using ALD deposition. The electrical conductivity of the film increased linearly with increasing duration of the ZnO deposition cycle (thickness), whereas other physical characteristics, such as surface energy and roughness, tended to saturate at a certain value. Differences in conductivity dramatically affected the behavior of SF295 glioblastoma cells grown on ZnO films, with high conductivity (thick) ZnO films causing growth arrest and producing SF295 cell morphologies distinct from those cultured on insulating substrates. Based on simple electrostatic calculations, we propose that cells grown on highly conductive substrates may strongly adhere to the substrate without focal-adhesion complex formation, owing to the enhanced electrostatic interaction between cells and the substrate. Thus, the inactivation of focal adhesions leads to cell proliferation arrest. Taken together, the work presented here confirms that substrates with high conductivity disturb the cell-substrate interaction, producing cascading effects on cellular morphogenesis and disrupting proliferation, and suggests that ALD-grown ZnO offers a single-variable method for uniquely tailoring conductivity.

Micropillar arrays as potential drug screens: Inhibition of micropillar-mediated activation of the FAK-Src-paxillin signaling pathway by the CK2 inhibitor CX-4945.

UNLABELLED Here, we demonstrate the possible applications of micropillar arrays in screening anti-metastasis drugs. Human lung adenocarcinoma A549 cells incubated in multiwell plates containing micropillars exhibited markedly different physical/biochemical behavior depending on pillar dimensions. In particular, A549 cells grown in plates containing 2-μm diameter, 16-μm pitched pillar arrays showed epithelial-to-mesenchymal transition (EMT)-like behavior; cell body elongation, and highly increased activation of the focal adhesion kinase (FAK)-Src-paxillin signaling cascade. FAK is the most prominent kinase involved in dynamic regulation of the actin cytoskeleton and cell adhesion, migration, and invasion. Activation of FAK, a hallmark of cancer cell adhesion and migration, is normally induced by various growth factors, such as transforming growth factor-β (TGF-β). Here, we found that pillar-mediated activation of signaling molecules mimicked that induced by TGF-β. Notably, micropillar arrays with specific dimensions accelerated the elongation of cells, an effect linked to the activation of signaling molecules related to EMT. Micropillar-induced FAK activation could be arrested by the casein kinase-2 (CK2) inhibitor CX-4945, a drug candidate with activity against TGF-β-induced cancer cell metastasis, demonstrating the possibility of using inorganic microstructures for cell-based drug screening. STATEMENT OF SIGNIFICANCE In this work, we have fabricated flexible substrates with regular arrays of micrometersized pillars, and used them to grow A549 human lung adenocarcinoma cells. Cells exhibit dramatically different behavior depending on the intervals of pillars. Especially, cells grown in certain pillar structures show epithelial-to mesenchmal transition (EMT)-like morphology and related molecules, which is similar to the activation obtained using expensive cytokine TGF-β. Based on the fact that pillar arrays may activate EMT like transition, screening of anti-cancer drug using pillar arrays have demonstrated as well in our work. Our study confirms that mechanical stimulation may exert similar effects with chemical stimulation, and such mechanical structures could be used as a large-scale drug screening platforms. Cell morphogenesis on engineered substrate is not new, but the present work could be distinguished with its unique fabrication process that can mass produce the structures and it could be applied for high-throughput drug screening. Also, we suggest the formation of focal adhesions on pillar structures and consequent strain as the possible mechanism behind the observed EMT-like transition. Currently, we are working on full-scale profiling of metabolomics and proteomics of cells grown in large-scale pillar arrays as well.

Performance assessments of vertically aligned carbon nanotubes multi-electrode arrays using Cath.a-differentiated (CAD) cells

In this work, Cath.a-differentiated (CAD) cells were used in place of primary neuronal cells to assess the performance of vertically aligned carbon nanotubes (VACNTs) multi-electrode arrays (MEA). To fabricate high-performance MEA, VACNTs were directly grown on graphene/Pt electrodes via plasma enhanced chemical deposition technique. Here, graphene served as an intermediate layer lowering contact resistance between VACNTs and Pt electrode. In order to lower the electrode impedance and to enhance the cell adhesion, VACNTs-MEAs were treated with UV-ozone for 20 min. Impedance of VACNTs electrode at 1 kHz frequency exhibits a reasonable value (110 kΩ) for extracellular signal recording, and the signal to noise ratio the is good enough to measure low signal amplitude (15.7). Spontaneous firing events from CAD cells were successfully measured with VACNTs MEAs that were also found to be surprisingly robust toward the biological interactions.

Ramp Activity Expert System for Scheduling and Coordination at an Airport

In this project, we have developed the ramp activity coordination expert system (races) to solve aircraft-parking problems. races includes a knowledge-based scheduling system that assigns all daily arriving and departing flights to the gates and remote spots with domain-specific knowledge and heuristics acquired from human experts. races processes complex scheduling problems such as dynamic interrelations among the characteristics of remote spots-gates and aircraft with various other constraints, for example, customs and ground-handling factors, at an airport. By user-driven modeling for end users and near-optimal knowledge-driven scheduling acquired from human experts, races can produce parking schedules for about 400 daily flights in approximately 20 seconds; human experts normally take 4 to 5 hours to do the same. Scheduling results in the form of Gantt charts produced by races are also accepted by the domain experts. races is also designed to deal with the partial adjustment of the schedule when unexpected events occur. After daily scheduling is completed, the messages for aircraft change, and delay messages are reflected and updated into the schedule according to the knowledge of the domain experts. By analyzing the knowledge model of the domain expert, the reactive scheduling steps are effectively represented as the rules, and the scenarios of the graphic user interfaces are designed. Because the modification of the aircraft dispositions, such as aircraft changes and cancellations of flights, is reflected in the current schedule, the modification should be sent to races from the mainframe for the reactive scheduling. The adjustments of the schedule are made semiautomatically by races because there are many irregularities in dealing with the partial rescheduling.

Rapid Synthesis of AgInS 2 /ZnS Core/Shell Nanoparticles and Their Luminescence Property

We have successfully synthesized core and /ZnS core/shell nanoparticles by the sonochemical method. The ultrasonic based and /ZnS nanoparticle synthesis can be utilized as a simple and rapid method. The /ZnS nanoparticles show the higher fluorescence intensity and quantum yield than nanoparticles. Fluorescence wavelength of /ZnS shows blue shift from 635 nm to 610 nm against because of reducing the defect sites and increasing spatial confinements. For the fluorescence lifetime, /ZnS (124.8 ns) has longer lifetime than (54.8 ns).

Shared Tag RFID System for Multiple Application Objects

Recently, the information technology has evolved toward the ubiquitous environment accessible to the network everywhere and every time. The ubiquitous environment can provide easy access to the devices and make one’s economical benefit also. The RFID system is an important core technology in ubiquitous environment. RFID system consists of contactless devices to communicate each other by radio frequency. It provides technologies of automatic object identification in invisible range, read/write function and adaptability against various circumstances. These advantages make RFID system to be applied in various fields and expect one of the big markets in the area of human life such as traffic card system, toll gate system, logistics, access control, etc. As RFID tags identify many different types of objects, it is going to increase the tags that have to be carry in individual life. But it is uneasy for a person to control many tags in a hand because traditional RFID systems have restriction that is one tag per each object and it is difficult to distinguish tags without some kind of effort. That is why a tag is used to store identifying information just for a single object in common RFID applications. We propose a multiple objects tag structure which can be shared by many different applications. As a tag is used to identify only one type of object, it is expected to have many tags increasingly by people or things. If a tag can be shared by many RFID application objects, it will be more efficient to RFID users and will be helpful to the information integration as well as device sharing. So, we design a RFID tag structure which has many different identifiers. This tag can be used to access many different RFID applications. We also propose an efficient authentication protocol adapted to the multiple objects tag structure. We consider robustness of the authentication protocol against various attacks in the proposed scheme. RFID system often makes serious violation of privacy and security caused by various attacks through the weak wireless interface. Eavesdropping, location tracking, spoofing, message losses or replay attack can threaten RFID components anywhere and anytime. To protect RFID system from these kinds of attacks, researchers have studied several schemes such as Faraday cage scheme (mCloak, 2003), blocker-tag scheme (Juels et al., 2003), hash lock scheme (Weis et al., 2003), randomized hash lock (Weis et al., 2003), hash chain (Ohkubo et al., 2003) and variable ID scheme (Saito & Sakurai, 2003), etc. However, these schemes have restrictions that each object is just corresponding to one tag. So, it is O pe n A cc es s D at ab as e w w w .in te ch w eb .o rg