Search

For several years, a growing interest among numerous researchers and investors in predicting stock price movements has spurred extensive exploration into employing advanced deep learning models. These models aim to develop systems capable of comprehending the stock market's complex nature. Despite the immense challenge posed by the diverse factors influencing stock price forecasting, there remains a notable lack of research focused on identifying the essential feature set for accurate predictions. In this study, we propose a Dynamic Feature Selection System (DFSS) to predict stock prices across the 10 major industries, as classified by the FnGuide Industry Classification Standard (FICS) in South Korea. We apply 16 feature selection algorithms from filter, wrapper, embedded, and ensemble categories. Subsequently, we adjust the settings of industry-specific index data to evaluate the model's performance and robustness over time. Our comprehensive results identify the optimal feature sets that significantly impact stock prices within each sector at specific points in time. By analyzing the inclusion ratios and significance of the optimal feature set by category, we gain insights into the proportion of feature classes and their importance. This analysis ensures the interpretability and reliability of our model. The proposed methodology complements existing methods that do not consider changes in the types of variables significantly affecting stock prices over time by dynamically adjusting the input variables used for learning. The primary goal of this study is to enhance active investment strategies by facilitating the creation of diversified portfolios for individual stocks across various sectors, offering robust models and feature sets that consistently demonstrate high performance across industries over time. [ABSTRACT FROM AUTHOR]

We investigated the impact of CaCO3 addition on the density and compressive strength of calcium aluminate cement (CAC)-based cementitious materials in binder jetting additive manufacturing (BJAM). To confirm the formation of a uniform powder bed, we examined the powder flowability and powder bed density for CaCO3 contents ranging from 0 to 20 wt.%. Specifically, powders with avalanche angles between 40.1–45.6° formed a uniform powder bed density with a standard deviation within 1%. Thus, a 3D printing specimen (green body) fabricated via BJAM exhibited dimensional accuracy of less than 1% across the entire plane. Additionally, we measured the hydration characteristics of CAC and the changes in compressive strength over 30 days with the addition of CaCO3. The results indicate that the addition of CaCO3 to CAC-based cementitious materials forms multimodal powders that enhance the density of both the powder bed and the green body. Furthermore, CaCO3 promotes the formation of highly crystalline monocarbonate (C4AcH11) and stable hydrate (C3AH6), effectively inhibiting the conversion of CAC and showing compressive strengths of up to 5.2 MPa. These findings suggest a strong potential for expanding the use of BJAM across various applications, including complex casting molds, cores, catalyst supports, and functional architectural interiors. [ABSTRACT FROM AUTHOR]

Highly selective etching of silicon nitride (Si3N4) and silicon dioxide (SiO2) has received considerable attention from the semiconductor community owing to its precise patterning and cost efficiency. We investigated the etching selectivity of Si3N4 and SiO2 in an NF3/O2 radio-frequency glow discharge. The etch rate linearly depended on the source and bias powers, whereas the etch selectivity was affected by the power and ratio of the gas mixture. We found that the selectivity can be controlled by lowering the power with a suitable gas ratio, which affects the surface reaction during the etching process. X-ray photoelectron spectroscopy of the Si3N4 and QMS measurements support the effect of surface reaction on the selectivity change by surface oxidation and nitrogen reduction with the increasing flow of O2. We suggest that the creation of SiOxNy bonds on the surface by NO oxidation is the key mechanism to change the etch selectivity of Si3N4 over SiO2. [ABSTRACT FROM AUTHOR]

In this study, the effects of an Al2O3 interfacial layer (IL) on the characteristics of ferroelectric field-effect transistors (FeFETs) with Hf0.5Zr0.5O2 (HZO) gate dielectrics on Si substrates were investigated. FeFETs with HZO gate dielectrics have gained considerable attention owing to their compatibility with modern fabrication processes and scalability. However, during HZO deposition on Si substrates, an ultrathin metal silicate IL with a low dielectric constant is formed in an uncontrolled manner, leading to a significant voltage drop and the generation of interface traps during device operation. To address this issue, an ultrathin Al2O3 IL with a thickness less than 2 nm was introduced between the HZO film and Si substrate via in situ atomic layer deposition. The impact of this IL on a memory window (MW) and endurance characteristics was evaluated by comparing the devices with and without an intentional Al2O3 IL. The obtained results revealed that the Al2O3 IL effectively suppressed the interface trap generation, expanded the MW, and enhanced the transistor endurance characteristics. This described approach can be potentially used for improving the reliability of FeFETs fabricated on Si substrates. [ABSTRACT FROM AUTHOR]