Search Results (63)

Modeling high-frequency volatility is an important topic of market microstructure, as it provides the empirical tools to measure and analyze the rapid price movements. Yet, volatility at a high frequency often exhibits abrupt shifts driven by news and trading activity, making accurate estimation challenging. This study develops a change-point duration (CPD) model to estimate spot volatility, in which price-change intensities remain constant between events but may shift at random change points. Using simulations and empirical analysis of Nasdaq limit order book data, we demonstrate that the CPD model achieves a favorable balance between responsiveness to sudden shocks and stability in volatility dynamics. Moreover, it outperforms benchmark approaches, including the classical autoregressive conditional duration model, nonparametric duration-based estimators, and candlestick-based measures. These findings highlight the CPD framework as an effective tool for volatility estimation in high-frequency trading environments. Full article

To address the triple bottlenecks of data scarcity, oversized models, and slow inference that hinder Cantonese automatic speech recognition (ASR) in low-resource and edge-deployment settings, this study proposes a cost-effective Cantonese ASR system based on LoRA fine-tuning and INT8 quantization. First, Whisper-tiny is parameter-efficiently fine-tuned on the Common Voice zh-HK training set using LoRA with rank = 8. Only 1.6% of the original weights are updated, reducing the character error rate (CER) from 49.5% to 11.1%, a performance close to full fine-tuning (10.3%), while cutting the training memory footprint and computational cost by approximately one order of magnitude. Next, the fine-tuned model is compressed into a 60 MB INT8 checkpoint via dynamic quantization in ONNX Runtime. On a MacBook Pro M1 Max CPU, the quantized model achieves an RTF = 0.20 (offline inference 5 × real-time) and 43% lower latency than the FP16 baseline; on an NVIDIA A10 GPU, it reaches RTF = 0.06, meeting the requirements of high-concurrency cloud services. Ablation studies confirm that the LoRA-INT8 configuration offers the best trade-off among accuracy, speed, and model size. Limitations include the absence of spontaneous-speech noise data, extreme-hardware validation, and adaptive LoRA structure optimization. Future work will incorporate large-scale self-supervised pre-training, tone-aware loss functions, AdaLoRA architecture search, and INT4/NPU quantization, and will establish an mJ/char energy–accuracy curve. The ultimate goal is to achieve CER ≤ 8%, RTF < 0.1, and mJ/char < 1 for low-power real-time Cantonese ASR in practical IoT scenarios. Full article

To address the limitations of existing knowledge graph-based recommendation algorithms, including insufficient utilization of semantic information and inadequate modeling of user behavior motivations, we propose SKGRec, a novel recommendation model that integrates knowledge graph and semantic features. The model constructs a semantic interaction graph (USIG) of user behaviors and employs a self-attention mechanism and a ranked optimization loss function to mine user interactions in fine-grained semantic associations. A relationship-aware aggregation module is designed to dynamically integrate higher-order relational features in the knowledge graph through the attention scoring function. In addition, a multi-hop relational path inference mechanism is introduced to capture long-distance dependencies to improve the depth of user interest modeling. Experiments on the Amazon-Book and Last-FM datasets show that SKGRec significantly outperforms several state-of-the-art recommendation algorithms on the Recall@20 and NDCG@20 metrics. Comparison experiments validate the effectiveness of semantic analysis of user behavior and multi-hop path inference, while cold-start experiments further confirm the robustness of the model in sparse-data scenarios. This study provides a new optimization approach for knowledge graph and semantic-driven recommendation systems, enabling more accurate capture of user preferences and alleviating the problem of noise interference. Full article

This study investigates how the rise of commission-free FinTech platforms and the introduction of fractional trading (FT) have altered trading behavior and order book dynamics in the NASDAQ equity market. Leveraging high-frequency ITCH data from highly capitalized stocks—AAPL, AMZN, GOOG, and TSLA—we analyze market microstructure changes surrounding the implementation of FT. Our empirical findings show a statistically significant increase in price levels, average tick sizes, and price volatility in the post-FinTech-FT period, alongside elevated price impact factors (PIFs), indicating steeper and less liquid limit order books. These shifts reflect greater participation by non-professional investors with limited order placement precision, contributing to noisier price discovery and heightened intraday risk. The altered liquidity landscape and increased volatility raise important questions about the resilience and informational efficiency of modern equity markets under democratized access. Our findings contribute to the growing literature on retail trading and provide actionable insights for market regulators and exchanges evaluating the design and oversight of evolving trading mechanisms. Full article

The Holy Bible is the most read book in the world, originally written in Aramaic, Hebrew, and Greek over a time span in the order of centuries by many people, and formed by a combination of various literary styles, such as stories, prophecies, poetry, instructions, and others. As such, the Bible is a complex text to be analyzed by humans and machines. This paper provides a systematic survey of the application of Artificial Intelligence (AI) and some of its subareas to the analysis of the Biblical scriptures. Emphasis is given to what types of tasks are being solved, what are the main AI algorithms used, and their limitations. The findings deliver a general perspective on how this field is being developed, along with its limitations and gaps. This research follows a procedure based on three steps: planning (defining the review protocol), conducting (performing the survey), and reporting (formatting the report). The results obtained show there are seven main tasks solved by AI in the Bible analysis: machine translation, authorship identification, part of speech tagging (PoS tagging), semantic annotation, clustering, categorization, and Biblical interpretation. Also, the classes of AI techniques with better performance when applied to Biblical text research are machine learning, neural networks, and deep learning. The main challenges in the field involve the nature and style of the language used in the Bible, among others. Full article

In recent years, high-frequency trading has become increasingly popular in financial markets, making the dynamic modeling of the limit book and the optimization of market maker strategies become key topics. However, existing studies often lacked detailed descriptions of order books and failed to fully characterize the optimal decisions of market makers in complex market environments, especially in China’s A-share market. Based on Markov queue theory, this paper proposes the dynamic model of the limit order and the optimal strategy of the market maker. The model uses a state transition probability matrix to refine the market diffusion state, order generation, and trading process and incorporates indicators such as optimal quote deviation and restricted order trading probability. Then, the optimal control model is constructed and the reference strategy is derived using the Hamilton–Jacobi–Bellman (HJB) equation. Then, the key parameters are estimated using the high-frequency data of Ping An Bank for a single trading day. In the empirical aspect, the six-month high-frequency trading data of 114 representative stocks in different market states such as the bull market and bear market in China’s A-share market were selected for strategy verification. The results showed that the proposed strategy had robust returns and stable profits in the bull market and that frequent capture of market fluctuations in the bear market can earn relatively high returns while maintaining 50% of the order coverage rate and 66% of the stable order winning rate. Our study used Markov queuing theory to describe the state and price dynamics of the limit order book in detail and used optimization methods to construct and solve the optimal market maker strategy. The empirical aspect broadens the empirical scope of market maker strategies in the Chinese market and studies the stability and effectiveness of market makers in different market states. Full article

We develop a deep reinforcement learning (RL) framework for an optimal market-making (MM) trading problem, specifically focusing on price processes with semi-Markov and Hawkes Jump-Diffusion dynamics. We begin by discussing the basics of RL and the deep RL framework used; we deployed the state-of-the-art Soft Actor–Critic (SAC) algorithm for the deep learning part. The SAC algorithm is an off-policy entropy maximization algorithm more suitable for tackling complex, high-dimensional problems with continuous state and action spaces, like those in optimal market-making (MM). We introduce the optimal MM problem considered, where we detail all the deterministic and stochastic processes that go into setting up an environment to simulate this strategy. Here, we also provide an in-depth overview of the jump-diffusion pricing dynamics used and our method for dealing with adverse selection within the limit order book, and we highlight the working parts of our optimization problem. Next, we discuss the training and testing results, where we provide visuals of how important deterministic and stochastic processes such as the bid/ask prices, trade executions, inventory, and the reward function evolved. Our study includes an analysis of simulated and real data. We include a discussion on the limitations of these results, which are important points for most diffusion style models in this setting. Full article

With the recent increase in retail investor participation led by commission-less fintech trading applications and new features like fractional trading, we now have higher volatility and significantly quicker price changes. This makes it hard to make informed trading decisions. Moreover, these effects are exacerbated even further around earnings announcements days. In this paper, we use Nasdaq data feed at a minute frequency and show that there is a significant increase in the slope of the price–volume structure during extended hours (after-hours, or pre-market hours) as compared with the ones observed during regular market times. Our analysis shows that the liquidity is much less during the extended market hours. As such, earnings announcements of stocks during these times have a significantly larger price impact than those stocks that have their earnings announced during regular trading hours. This significant difference can be explained by observing the limit order book structures during these different trading periods. We suggest that the earnings announcements should not be made during extended hours given the significantly lower liquidity and thus, the significantly larger price impact that not only determines the prices for the next trading session but also sets the new “fundamental” price signals for the stocks. Full article

This study examines the intraday dynamics of liquidity and trading activity on the Egyptian Exchange (EGX) to assess its market quality. Using reconstructed five-minute limit order book data, this study measures liquidity dimensions and explores anomalies through interval-of-day and day-of-week models. Key findings reveal an inverted J-shaped pattern in spreads due to information asymmetry, a U-shaped pattern in total depth, and a J-shaped market depth pattern. Additionally, significant day-of-week effects are observed, with Sundays showing the lowest liquidity and Thursdays the highest trading activity. These patterns highlight the impact of the EGX’s unique microstructure, including tick sizes and a preference for limit orders. This study underscores the influence of market structure on liquidity, trading efficiency, and cost, emphasizing the need for tailored regulatory and trading strategies. It provides valuable insights for investors optimizing trading strategies and policymakers seeking to enhance market integrity. Concluding, this research offers a foundation for understanding intraday liquidity patterns in emerging markets like the EGX and proposes future exploration of how information flows and trading mechanisms affect price discovery and market efficiency. Full article

Background and Objectives: This systematic review aimed to compare the effect of chitosan in smear layer removal with other commonly used chelators during root canal treatment. Materials and Methods: The PRISMA guidelines were followed. Ex vivo studies performed in non-endodontically treated extracted human permanent teeth with a fully formed apex, in which sodium hypochlorite was the main irrigant and chitosan was used as final irrigation to observe its capacity to remove the smear layer using a Scanning Electron Microscope (SEM), were included. In addition, reviews, letters, opinion articles, conference abstracts, book chapters, or articles that did not use a control group were excluded. A literature search was undertaken without limits on time or language, until February 2024, in PubMed—MEDLINE, Scopus, Web of Science, and in the electronic archives of four endodontic journals. The risk of bias was evaluated by adapting the risk of bias assessment used in a previous study. Study selection, data collection, and synthesis were performed and the risk of bias was assessed by two independent reviewers. Results: Six studies fulfilled the eligibility criteria and were included. Four studies found chitosan to be as effective as EDTA and one paper showed it was more effective than EDTA and MTAD; however, one article found it to be comparable to citric acid. The overall risk of bias was medium. Quantitative analysis of the results was not possible due to the heterogeneity found between the study methodologies of the included articles. Conclusions: Within the limitations of this study, 0.2% chitosan may be considered as a promising irrigation solution when employed as a final irrigant in order to remove the smear layer. Nonetheless, a standardized protocol for the use of chelators in root canal treatment should be established in future studies. Full article

We investigate the common movement and information content of trades at steps away from the best bid and offer (BBO) using Tokyo Stock Exchange data. We create strings, a series of trades at the same or at an inferior price. The number of the strings is invariant for securities across trading days. The number of shares traded during a string and the time needed for the completion of a string are also significantly related across days for a given stock. The strings represent liquidity beyond the BBO. In addition, the strings characterize the price adjustment process in which we relate to the information on the underlying asset value. The strings measure order aggressiveness beyond the BBO. Finally, we show that the return for the strings is significantly related to the state of the limit order book at the start of the string. Thus, traders can infer information using strings to achieve higher returns. Full article

Myrrh is the resinous substance secreted by plants of the genus Commiphora. In traditional Chinese medicine, Ayurvedic medicine, and traditional Arabic medicine, myrrh is regarded as an important medicinal material, widely used in the treatment of trauma, arthritis, hyperlipidemia, and other diseases. This review explores the evolving scientific understanding of the genus Commiphora, covering facets of ethnopharmacology, phytochemistry, pharmacology, artificial cultivation, and quality control. In particular, the chemical constituents and pharmacological research are reviewed. More than 300 types of secondary metabolites have been identified through phytochemical studies of this genus. Guggulsterone is a bioactive steroid isolated mainly from Commiphora mukul. The two isomers, Z- and E-guggulsterone, have shown a wide range of in vitro and in vivo pharmacological effects, including anti-proliferation, antioxidant, anti-inflammatory, and antibacterial. However, the current scientific research on quality control of medicinal materials and identification of original plants is insufficient, which limits the reproducibility and accuracy of biological activity evaluation experiments. Therefore, the establishment of analytical protocols and standardization of extracts is an important step before biological evaluation. At the same time, in order to find more bioactive substances, it is necessary to strengthen the research on the stems, barks, and leaves of this genus. The sources used in this study include PubMed, CNKI, Web of Science, Google Scholar, and other databases, as well as multinational pharmacopoeias, ancient books of traditional medicine, herbal classics, and modern monographs. Full article

Background: Although computational models are advancing air quality prediction, achieving the desired performance or accuracy of prediction remains a gap, which impacts the implementation of machine learning (ML) air quality prediction models. Several models have been employed and some hybridized to enhance air quality and air quality index predictions. The objective of this paper is to systematically review machine and deep learning techniques for spatiotemporal air prediction challenges. Methods: In this review, a methodological framework based on PRISMA flow was utilized in which the initial search terms were defined to guide the literature search strategy in online data sources (Scopus and Google Scholar). The inclusion criteria are articles published in the English language, document type (articles and conference papers), and source type (journal and conference proceedings). The exclusion criteria are book series and books. The authors’ search strategy was complemented with ChatGPT-generated keywords to reduce the risk of bias. Report synthesis was achieved by keyword grouping using Microsoft Excel, leading to keyword sorting in ascending order for easy identification of similar and dissimilar keywords. Three independent researchers were used in this research to avoid bias in data collection and synthesis. Articles were retrieved on 27 July 2024. Results: Out of 374 articles, 80 were selected as they were in line with the scope of the study. The review identified the combination of a machine learning technique and deep learning techniques for data limitations and processing of the nonlinear characteristics of air pollutants. ML models, such as random forest, and decision tree classifier were among the commonly used models for air quality index and air quality predictions, with promising performance results. Deep learning models are promising due to the hyper-parameter components, which consist of activation functions suitable for nonlinear spatiotemporal data. The emergence of low-cost devices for data limitations is highlighted, in addition to the use of transfer learning and federated learning models. Again, it is highlighted that military activities and fires impact the O₃ concentration, and the best-performing models highlighted in this review could be helpful in developing predictive models for air quality prediction in areas with heavy military activities. Limitation: This review acknowledges methodological challenges in terms of data collection sources, as there are equally relevant materials on other online data sources. Again, the choice and use of keywords for the initial search and the creation of subsequent filter keywords limit the collection of other relevant research articles. Full article

Due to the imbalance between supply and demand, liner container transportation often faces the problem of low slot utilization, which will occur in the shipping process, such as dry container demand exceeding the available dry slots and reefer slots not being fully utilized. This makes it important and challenging to maintain a balance between the actual demand and the limited number of slots allocated for liner container transport. Therefore, this study proposes a flexible allocation method: expanding the types of containers that can be loaded in the same slot. This method is suitable for handling each dynamic arrival container booking request by shipping enterprises, making decisions to accept or reject, and flexibly allocating shipping slots. In order to maximize the total revenue generated by accepting container booking requests during the entire booking acceptance cycle, we establish a dynamic programming model for the flexible allocation of slots. For model solving, we use the Q-learning reinforcement learning algorithm. Compared with traditional heuristic algorithms, this algorithm can improve solving efficiency and facilitate decision-making at the operational level of shipping enterprises. In terms of model performance, examples of different scales are used for comparison and training; the results are compared with the model without flexible allocation, and it is proved that the model proposed in this paper can obtain higher returns than the model without flexible allocation. The results show that the model and Q-learning algorithm can help enterprises solve the problem of the flexible allocation of shipping slots, and thus, this research has practical significance. Full article

Understanding how price-volume information determines future price movement is important for market makers who frequently place orders on both buy and sell sides, and for traders to split meta-orders to reduce price impact. Given the complex non-linear nature of the problem, we consider the prediction of the movement direction of the mid-price on an option order book, using machine learning tools. The applicability of such tools on the options market is currently missing. On an intraday tick-level dataset of options on an exchange traded fund from the Chinese market, we apply a variety of machine learning methods, including decision tree, random forest, logistic regression, and long short-term memory neural network. As machine learning models become more complex, they can extract deeper hidden relationship from input features, which classic market microstructure models struggle to deal with. We discover that the price movement is predictable, deep neural networks with time-lagged features perform better than all other simpler models, and this ability is universal and shared across assets. Using an interpretable model-agnostic tool, we find that the first two levels of features are the most important for prediction. The findings of this article encourage researchers as well as practitioners to explore more sophisticated models and use more relevant features. Full article