Information Technology Institute

A comparative study of particle defects on cleaned GaSb wafers with respect to their morphology, elemental composition, and origin have been investigated. Extensive range of advanced characterization techniques including KLA candela, time-of-flight secondary ion mass spectrometry, and scanning electron microscopy and energy dispersive x-ray spectroscopy have been used to give the results of particle distribution and chemical composition. The results indicate that the distribution and composition of residual particles exhibit significant complexity, rather than the particles themselves. The particle defects on the surface of GaSb are mainly the accumulation and adsorption of organic and inorganic substances used in the polishing and cleaning processes. In-depth analysis of particle properties and origins is conducted, and the incorporation of appropriate surfactants effectively reduced the particle count on the substrate surface. This work offers a solution for enhancing surface quality and facilitating the production of high-performance devices.

The goal of this paper is to combine algebraic quantum field theory with methods from modal logic and the causal set program. We introduce a factorially damped past-influence operator on causal sets, formulate locally covariant nets over Alexandrov intervals, and give discrete analogues of the Haag–Kastler axioms. We close by sketching a route toward dynamics via sequential growth and incidence-algebra localization.

Abstract Detecting fraudulent trading activities in the FinTech sector presents significant challenges due to the highly dynamic nature of financial markets and the rapid adaptation strategies employed by fraudsters. Conventional machine learning techniques, especially when trained on imbalanced datasets, often exhibit classification bias toward the majority class (legitimate transactions), reducing their ability to identify fraudulent behaviors. To address this limitation, we introduce a hybrid memetic-ABC framework that combines Artificial Bee Colony (ABC) sampling for global exploration with memetic local refinement to generate high-quality synthetic minority-class samples guided by feature importance. ABC ensures broad coverage of potential fraudulent patterns, while the memetic refinement fine-tunes synthetic samples and feature subsets to enhance discriminative power, capturing subtle anomalies in transaction amounts, locations, and timing. The framework further incorporates an anomaly detection module that constructs behavioral profiles for traders using historical transaction data, clustering them based on similar trading patterns. This enables identification of key indicators of potential fraud, such as sudden volume fluctuations, irregular transaction timings, and deviations from behavioral norms. For real-time monitoring and scalable deployment in decentralized environments, the system is extended into a Distributed Intrusion Detection System (DIDS) architecture, simultaneously detecting transactional anomalies and network-level threats, including DDoS attacks and unauthorized access attempts. By correlating anomalous trading patterns with DIDS-generated alerts, the approach achieves enhanced detection accuracy and robustness against complex fraud scenarios. Experimental evaluations demonstrate that the proposed model outperforms state-of-the-art techniques, achieving approximately 10% improvement in accuracy.

The rapid proliferation of misinformation across online platforms poses a challenge that calls for advanced real-time detection and verification mechanisms. Most existing methods depend on some form of user inspection or centralized system which tends to be slow, biased, and subjective. In this paper, we address these gaps by integrating Natural Language Processing (NLP), Machine Learning (ML), and blockchain technology in a single novel framework. The system functions through three primary processes: event detection, trust evaluation, and validation through blockchain. In the primary step, NLP models examine data streams like news articles, tweets, and forum posts to capture major events across diverse areas including politics, finance, healthcare, and natural disasters. The second step applies ML algorithms to evaluate the identified events based on source credibility, narrative reliability, and factual consistency analysis. In the last stage, the verification process is facilitated through the use of blockchain technology which assures transparency, immutability, decentralization, and security by recording verified events alongside their trust scores on a blockchain ledger. This multi-disciplinary approach not only enhances accuracy and boosts accountability but also streamlines real-time processing while tackling the challenges of trusted event detection and verification. By applying NLP, ML, and blockchain, this paper proposes a new paradigm of trust enhancement for digital information.

Detecting online fraudulent trading in the realm of Fintech presents several challenges, primarily due to the dynamic nature of financial markets and the evolving tactics of fraudsters. Traditional machine learning algorithms trained on unbalanced datasets tend to bias towards the majority class (legitimate transactions) and may overlook minority class (fraudulent transactions) patterns. This bias can lead to poor performance in detecting fraudulent activities. The choice of sampling technique (e.g., oversampling, undersampling, SMOTE) can significantly impact model performance. However, selecting the appropriate sampling strategy requires domain knowledge and experimentation, which can be time-consuming and resource-intensive. This work utilizes Artificial Bee Colony (ABC)-based sampling to tackle class imbalance in credit card fraud detection. By generating realistic synthetic fraud samples, ABC-sampling helps the model learn fraudulent patterns more effectively without favoring non-fraudulent transactions. Inspired by the foraging behavior of bees, the process involves exploring existing fraud patterns, selecting the most relevant ones, creating synthetic fraud samples, and refining them to ensure they closely resemble real fraud cases while preserving key features that distinguish fraud from regular transactions. This method enhances fraud detection accuracy and minimizes false alarms, making the system more reliable. The suggested model employs anomaly detection algorithm to identify unusual or fraudulent trading activities in which it creates behavioral profiles for individual traders based on their historical trading activities and utilizes machine learning algorithm to cluster traders into groups based on similar behavior patterns. Then it identifies characteristic features of fraudulent traders such as sudden changes in trading volume, irregular trading hours, or trading activities inconsistent with their profile. By analyzing patterns and anomalies in traders’ behavior, these approaches can effectively identify suspicious activities indicative of fraudulent behavior. Extensive performance studies demonstrate that the proposed algorithm significantly outperforms the state-of-the-art methods by 10% in terms of accuracy.

With the rapid growth of digital technology, social media has become a key platform for sharing information and opinions. Twitter, one of the most popular platforms in Indonesia, enables users to interact directly with public figures such as Najwa Shihab. This study aims to analyze public sentiment toward Najwa Shihab on Twitter using sentiment analysis, specifically employing the Naïve Bayes and Support Vector Machine (SVM) algorithms. Sentiment analysis is essential to understanding public opinion, as it classifies text into categories like positive, negative, or neutral, providing valuable insights into societal perspectives on public figures. In this study, 10,000 tweets related to Najwa Shihab were collected from January 1, 2023, to January 31, 2023. Data preprocessing steps such as data cleaning, tokenization, stopwords removal, and filtering were conducted to ensure high-quality data for analysis. The Naïve Bayes and SVM algorithms were applied using RapidMiner to classify the sentiment of the tweets. The performance of both algorithms was evaluated based on accuracy, precision, recall, and F1-score.The results revealed that SVM outperformed Naïve Bayes in all metrics, demonstrating its superior ability to classify sentiments correctly. The sentiment distribution indicated a majority of positive opinions toward Najwa Shihab, with fluctuations in negative sentiment during specific events. This study provides insights into public sentiment analysis and contributes to understanding social media opinions on public figures.

Transmission electron microscopy was employed to investigate the micromorphological properties of subsurface lattice damage (SLD) in ground and rough polished indium arsenide (InAs) substrates. We show that the types and distributions of SLD defects in ground substrates and rough polished substrates are different, obviously, and are more complex compared to those in other hard-brittle materials. Remarkably, the depth of the SLD layer containing severely damaged regions, abnormal contrasts, and moiré fringes in rough polished substrates is three times deeper than that of the SLD layer composed of dislocations, stacking faults, and subsurface cracks in ground substrates. SLD in the rough polished substrate exhibits defect characteristics of lattice glide caused by the force applied during the polishing process. Formation mechanisms of SLD during grinding and rough chemical-mechanical polishing processes of the InAs substrates are discussed in view of removal behaviors of surface material and the evolution of stress fields.

In zoom analog-to-digital converters (ADCs), due to the dynamic updating and scaling of fine reference voltages by coarse SAR ADC, traditional noise coupling technique cannot be directly applied to fine ΔΣ modulator, which may lead to modulator overload and a sharp decline in the signal-to-noise distortion ratio (SNDR). To address this issue, this article proposes a noise-coupled technique specifically designed for zoom ADCs. It establishes a mathematical model and determines the appropriate feedback voltage based on coarse quantization and the over-ranging factor, ultimately enhancing SNDR and energy efficiency. The zoom ADC is designed using a 180 nm CMOS process and incorporates two integrators to achieve 3rd-order noise-shaping capability (60 dB/dec). It achieves an SNDR of 105.7 dB at a sampling frequency of 256 kHz and an oversampling rate of 128, with a power consumption of only 90.25 μW. The SNDR-based Schreier figure-of-merit (FoMs) is 176.1 dB, indicating competitive performance in terms of high resolution and energy efficiency.

Abstract Among the various challenges in the field of organic light‐emitting diodes (OLEDs), simultaneously achieving high efficiency, a long lifespan, and a narrow full‐width at half maximum (FWHM) in blue OLEDs remains a significant hurdle. Herein, we demonstrate a strategy to improve the color purity of tetradentate Pt(II) complexes with the assistance of ⋅⋅⋅H interaction by incorporating trifluoromethyl (−CF3) groups into the well‐known blue tetradentate Pt(II) phosphorescent complex. The results show that the different substitution positions of −CF3 have significantly varying effects on the FWHM values of the complexes; specifically, introducing −CF3 on the benzene ring of carbazole effectively reduces the FWHM, while introducing it on the benzene ring linked to the carbene unit has a minimal impact. When utilized in a mixed host system of SiCzCz/SiTrzCz2, the OLEDs with these new complexes as emitters demonstrated maximum external quantum efficiencies (EQEs) of 27.1 %, 33.8 %, and 27.6 % for PtON‐CF3‐1, PtON‐CF3‐2, and PtON‐2CF3, respectively, with excellent color purity (CIEy≈0.14 for the PtON‐CF3‐2 based device). Notably, the device based on PtON‐CF3‐2 matched the stability of the benchmark PtON‐TBBI complex.This work offers an important guideline for the design of high efficient Pt(II) phosphorescent complexes with good blue color purity.