Search

User reviews such as SNS feeds and blog writings have been widely used to extract opinions, complains, and requirements about a given place or product from users’ perspective. However, during the process of collecting them, a lot of reviews that are irrelevant to a given search keyword can be included in the results. Such irrelevant reviews may lead to distorted results in data analysis. In this paper, we discuss a method to detect irrelevant user reviews efficiently by combining various oversampling and machine learning algorithms. About 35,000 user reviews collected from 25 restaurants and 33 tourist attractions in Ulsan Metropolitan City, South Korea, were used for learning, where the ratio of irrelevant reviews in the two kinds of data sets was 53.7% and 71.6%, respectively. To deal with skewness in the collected reviews, oversampling algorithms such as SMOTE, Borderline-SMOTE, and ADASYN were used. To build a model for the detection of irrelevant reviews, RNN, LSTM, GRU, and BERT were adopted and compared, as they are known to provide high accuracy in text processing. The performance of the detection models was examined through experiments, and the results showed that the BERT model presented the best performance, with an F1 score of 0.965.

The excellent multifunctional properties of magnetic nanoparticles (MNPs) allow them to be used in a wide range of applications. However, the required processes for the MNPs synthesis and characterization are still time-consuming and laborious, which further delays the research and development of the synthesis of MNPs. To overcome these difficulties, an on-chip magnetometer consisting of a planar Hall magnetoresistive (PHMR) sensor with a reactor was developed to optimize synthesis conditions for MNPs effectively. The microreactor was fabricated on the PHMR sensor, and the induced field from the synthesized magnetic particles was directly detected in the form of an output signal. Since the on-chip magnetometer enables simultaneous synthesis and measurement, it successfully monitored from the start point to the endpoint of the MNPs synthesis reaction. We studied the effect of the reagent ratio on the particle with iron chloride and aqueous ammonia solution, and we found the most suitable combination using a multi-array on-chip magnetometer. Experiments under various conditions were conducted simultaneously, and the experimental conditions were optimized in a short duration of time. Even though the synthesis reaction took place in a short time, the sensor's fast measurement rate and sensitivity allowed in-line measurement of the particle formation process. In addition, the developed on-chip magnetometer can be flexibly combined according to the desired configuration. Therefore, it is expected to be widely used as a new device for the simultaneous synthesis and analysis of MNPs.

Analysis of changes in particle trajectories due to collision is essential to understanding the physical information such as energy or mass exchange during collisions between particles. Most of the research has studied particles with initial energies in a relatively uniform or gradient static energy landscape. However, compared to the particle collisions of a static energy landscape, the movement becomes more diverse in a complex magnetic energy landscape. We designed micro-magnet arrays to create complex energy landscapes with temporal and spatial periodicity. Magnetic particles draw distinctly different trajectories in this magnetophoretic system depending on the coupling number. When the particles collide with each other, an immediate shift to a classified trajectory was confirmed according to the collision type. This study analyzes the particle trajectory shift that includes various information. Therefore, we inversely calculate the coupling history, such as the coupling number, time, and order due to collision. This investigation represents a unique analysis of the dynamics change of particles within a complex magnetic energy landscape.

Here, the authors present a pseudo-diamagnetophoresis mattertronic approach for programmable manipulation of label-free cells. Immersed in biocompatible ferrofluids, single cells are moved along linear negative micromagnetic patterns, switched at eclipse diode patterns and stored in potential wells.

Advances in micro- and nanotechnology have led to rapid employment of spintronic sensors in both recording and non-recording applications. These state-of-the-art magnetoresistive spintronic sensors exhibit high sensitivities and ultra-low field detectivities that meet requirements of smart sensing applications in the fields of internet of things, mobile devices, space technology, aeronautics, magnetic flux leakage, domotics, the environment, and healthcare. Moreover, their ability to be customized and miniaturized, ease of integration, and cost-effective nature make these sensors uniquely competitive with regard to mass production. In this study, we discuss magnetoresistive field sensors based on the planar-Hall effect, which are very promising for their high sensitivity and sensing ultra-low magnetic fields. We provide a detailed historical overview and discuss recent dramatic developments in several application fields. In addition, we discuss sensor material property requirements, design architectures, noise-reduction techniques, and sensing capabilities, along with the high repeatabilities and good flexibility characteristics of such devices. All of these high-performance characteristics apply across a wide temperature range and make the sensor robust when used in various novel applications. The sensor also appears promising because it is cost-effective and can be used in micro-sensing applications. Recently, a noteworthy study that combined integrated planar-Hall magnetoresistive sensors with microfluidic channels revealed their potential for highly localized magnetic field sensing. This characteristic makes them suitable for point-of-care-technologies that require resolutions of a few pT at room temperature.

As the number of cores equipped in a system grows, the impact of the spinlock waiting inside the operating system (OS) kernel on the performance and energy efficiency of a system worsens. In particular, it deteriorates the effectiveness of simultaneous multithreading (SMT). Because spinlocks are indispensable in OS kernels, it is necessary to suppress the spin wait overhead in the many-core systems. To address this issue, we propose the catnap spinlock that exploits the ACPI-C state, which is named as the catnap state and is induced by the MONITOR/MWAIT instruction pair. The catnap state releases the processor resources while deceiving the kernel that the thread is iterating a busy-waiting loop. Because entering and exiting from the C-state require considerable time, we applied the catnap loop only to the non-head waiters not to delay the lock handover operation. Furthermore, we selectively applied the catnap spinlock to the lock instances for sufficiently long critical sections based on the observation made in profiling runs. The proposed scheme was implemented in the Linux kernel and evaluated in a many-core processor system with a few workloads from the PARSEC and Re-aim benchmark suites. Our evaluation showed that the proposed scheme improved the performance by up to 33.59% and reduced energy consumption by 39.11%.

Abstract Cell clustering techniques are important to produce artificial cell clusters for in vitro models of intercellular mechanisms at the single‐cell level. The analyses considering physical variables such as the shape and size of cells have been very limited. In addition, the precise manipulation of cells and control of the physical variables are still challenging. In this paper, a magnetophoretic device consisting of a trampoline micromagnet and active elements that enable the control of individual selective jumping motion and positioning of a micro‐object is proposed. Based on a numerical simulation under various conditions, automatic separation or selective clustering of micro‐objects according to their sizes is performed by parallel control and programmable manipulation. This method provides efficient control of the physical variables of cells and grouping of cells with the desired size and number, which can be a milestone for a better understanding of the intercellular dynamics between clustered cells at the single‐cell level for future cell‐on‐chip applications.

The limited memory capacity of mobile devices leads to the popular use of compressed swap schemes, which reduce the I/O operations involving the swapping in and out of infrequently accessed pages. However, most of the current compressed swap schemes indiscriminately compress and store all swap-out pages. Considering that both energy and computing power are scarce resources in mobile devices, and modern applications frequently deal with already-compressed multimedia data, this blind approach may cause adverse impacts. In addition, they focus only on anonymous pages and not on file-mapped pages, because the latter are backed by on-disk files. However, our observations revealed that, in mobile devices, file-mapped pages consume significantly more memory than anonymous pages. Last but not least, most of the current compressed swap schemes blindly follow the least-recently-used (LRU) discipline when choosing the victim pages for replacement, not considering the compression ratio or data density of the cached pages. To overcome the aforementioned problems and maximize the memory efficiency, we propose a compressed swap scheme, called enhanced zswap (ezswap), for mobile devices. ezswap accommodates not only anonymous pages, but also clean file-mapped pages. It estimates the compression ratio of incoming pages with their information entropy, and selectively compresses and caches the pages only with beneficial compression ratios. In addition, its admission control and cache replacement algorithms are based on a cost-benefit model that considers not only the access recency of cached pages but also their information density and expected eviction cost. The proposed scheme was implemented in the Linux kernel for Android. Our evaluation with a series of commercial applications demonstrated that it reduced the amount of flash memory read by up to 55%, thereby improving the application launch time by up to 22% in comparison to the original zswap.

Abstract The manipulation of superparamagnetic beads has attracted various lab on a chip and magnetic tweezer platforms for separating, sorting, and labeling cells and bioentities, but the irreversible aggregation of beads owing to magnetic interactions has limited its actual functionality. Here, an efficient solution is developed for the disaggregation of magnetic beads and interparticle distance control with a magnetophoretic decoupler using an external rotating magnetic field. A unique magnetic potential energy distribution in the form of an asymmetric magnetic thin film around the gap is created and tuned in a controlled manner, regulated by the size ratio of the bead with a magnetic pattern. Hence, the aggregated beads are detached into single beads and transported in one direction in an array pattern. Furthermore, the simultaneous and accurate spacing control of multiple magnetic bead pairs is performed by adjusting the angle of the rotating magnetic field, which continuously changes the energy well associated with a specific shape of the magnetic patterns. This technique offers an advanced solution for the disaggregation and controlled manipulation of beads, can allow new possibilities for the enhanced functioning of lab on a chip and magnetic tweezers platforms for biological assays, intercellular interactions, and magnetic biochip systems.

The early removal of drug delivery agents before reaching the affected target remains an area of interest to researchers. Several magnetotactic bacteria (MTB) have been used as self-propelled drug delivery agents, and they can also be controlled by an external magnetic field. By attaching the PEG–biotin polymer, the bacteria are turned into a stealth material that can escape from the phagocytosis process and reach the area of interest with the drug load. In the study, we developed a potential drug carrier by attaching the PEG–biotin to the MTB-through-NHS crosslinker to form a MTB/PEG–biotin complex. The attachment stability, efficacy, and bacterial viability upon attachment of the PEG–biotin polymer were investigated. Biological applications were carried out using a cytotoxicity assay of THP-1 cells, and the results indicate that the MTB/PEG–biotin complex is less harmful to cell viability compared to MTB alone. Along with cytotoxicity, an assay for cell association was also evaluated to assess the complex as a potential stealth material. The development of these complexes focuses on an easy, time-saving, and stable technique of polymer attachment with the bacteria, without damaging the cell’s surface, so as to make it a strong and reliable delivery agent.

The resolution of planar-Hall magnetoresistive (PHMR) sensors was investigated in the frequency range from 0.5 Hz to 200 Hz in terms of its sensitivity, average noise level, and detectivity. Analysis of the sensor sensitivity and voltage noise response was performed by varying operational parameters such as sensor geometrical architectures, sensor configurations, sensing currents, and temperature. All the measurements of PHMR sensors were carried out under both constant current (CC) and constant voltage (CV) modes. In the present study, Barkhausen noise was revealed in 1/f noise component and found less significant in the PHMR sensor configuration. Under measured noise spectral density at optimized conditions, the best magnetic field detectivity was achieved better than 550 pT/√Hz at 100 Hz and close to 1.1 nT/√Hz at 10 Hz for a tri-layer multi-ring PHMR sensor in an unshielded environment. Furthermore, the promising feasibility and possible routes for further improvement of the sensor resolution are discussed.

Advanced microelectromechanical system (MEMS) magnetic field sensor applications demand ultra-high detectivity down to the low magnetic fields. To enhance the detection limit of the magnetic sensor, a resistance compensator integrated self-balanced bridge type sensor was devised for low-frequency noise reduction in the frequency range of 0.5 Hz to 200 Hz. The self-balanced bridge sensor was a NiFe (10 nm)/IrMn (10 nm) bilayer structure in the framework of planar Hall magnetoresistance (PHMR) technology. The proposed resistance compensator integrated with a self-bridge sensor architecture presented a compact and cheaper alternative to marketable MEMS MR sensors, adjusting the offset voltage compensation at the wafer level, and led to substantial improvement in the sensor noise level. Moreover, the sensor noise components of electronic and magnetic origin were identified by measuring the sensor noise spectral density as a function of temperature and operating power. The lowest achievable noise in this device architecture was estimated at ~3.34 nV/Hz at 100 Hz.

Magnetic sensors have great potential for biomedical applications, particularly, detection of magnetically-labeled biomolecules and cells. On the basis of the advantage of the planar Hall effect sensor, which consists of improved thermal stability as compared with other magnetic sensors, we have designed a portable biosensor platform that can detect magnetic labels without applying any external magnetic field. The trilayer sensor, with a composition of Ta (5 nm)/NiFe (10 nm)/Cu (x = 0 nm~1.2 nm)/IrMn (10 nm)/Ta (5 nm), was deposited on a silicon wafer using photolithography and a sputtering system, where the optimized sensor sensitivity was 6 μV/(Oe∙mA). The detection of the magnetic label was done by comparing the signals obtained in first harmonic AC mode (1f mode) using an external magnetic field and in the second harmonic AC mode (2f mode) with a self-field generated by current passing through the sensor. In addition, a technique for the β-amyloid biomarker-based antibody-antigen sandwich model was demonstrated for the detection of a series of concentrations of magnetic labels using the self-field mode method, where the signal-to-noise ratio (SNR) was high. The generated self-field was enough to detect an immobilized magnetic tag without an additional external magnetic field. Hence, it could be possible to reduce the device size to use the point-of-care testing using a portable circuit system.