Hasil untuk "Business mathematics. Commercial arithmetic. Including tables, etc."

Eliott Thomas, Mickael Coustaty, Aurelie Joseph et al.

Automating table extraction (TE) from business documents is critical for industrial workflows but remains challenging due to sparse annotations and error-prone multi-stage pipelines. While semi-supervised learning (SSL) can leverage unlabeled data, existing methods rely on confidence scores that poorly reflect extraction quality. We propose QUEST, a Quality-aware Semi-supervised Table extraction framework designed for business documents. QUEST introduces a novel quality assessment model that evaluates structural and contextual features of extracted tables, trained to predict F1 scores instead of relying on confidence metrics. This quality-aware approach guides pseudo-label selection during iterative SSL training, while diversity measures (DPP, Vendi score, IntDiv) mitigate confirmation bias. Experiments on a proprietary business dataset (1000 annotated + 10000 unannotated documents) show QUEST improves F1 from 64% to 74% and reduces empty predictions by 45% (from 12% to 6.5%). On the DocILE benchmark (600 annotated + 20000 unannotated documents), QUEST achieves a 50% F1 score (up from 42%) and reduces empty predictions by 19% (from 27% to 22%). The framework's interpretable quality assessments and robustness to annotation scarcity make it particularly suited for business documents, where structural consistency and data completeness are paramount.

Farnaz Fereidoonian, Seyed Jafar Sadjadi, Mehdi Heydari et al.

The issue of integrated scheduling and sequencing operation of unloading and loading equipment in container ports has been one of the most important issues concerning time efficiency. In addition, with the emergence of green harbor concepts, the inclusion of criteria for minimizing energy consumption, fuel and emission reduction are among the other issues that have been noticed by planners in the field of energy efficiency. Furthermore, due to the complexity and scope of activities of a container terminal, uncertainty in operational parameters such as transportation time, time of readiness and entry of work into the system and the velocity of the transportation fleet are inevitable in this operational environment. Therefore, this research with the aim of sharing trucks among loading and unloading equipment, proposes a robust multi-objective integer programming model for the synchronized scheduling of truck operations with other handling equipment to decrease the fuel consumption of trucks and the flow time of containers, considering the uncertainty in operational parameters as fuzzy numbers. To find the Pareto solutions for this model, the ε-Constraint technique is employed. Finally, the performance of the model in deterministic and uncertain modes is evaluated, compared and analyzed employing the inputs gathered from Shahid Rajaei port. The findings demonstrate that using this model will result in a substantial decrease in both fuel consumption and flow time of containers in comparison to the current procedure. Additionally, results will demonstrate the extent to which the terminal's fuel and time consumption will increase under conditions of uncertainty in operational parameters when the optimal plans derived from the robust model are implemented.

Cielo Gonzalez Moyano, Finn Klessascheck, Saimir Bala et al.

Business Process Management (BPM) is mostly centered around finding technical solutions. Nudging is an approach from psychology and behavioral economics to guide people's behavior. In this paper, we show how nudging can be integrated into the different phases of the BPM lifecycle. Further, we outline how nudging can be an alternative strategy for more sustainable business processes. We show how the integration of nudging offers significant opportunities for process mining and business process management in general to be more human-centric. We also discuss challenges that come with the adoption of nudging.

Zeping Yu, Sophia Ananiadou

We find arithmetic ability resides within a limited number of attention heads, with each head specializing in distinct operations. To delve into the reason, we introduce the Comparative Neuron Analysis (CNA) method, which identifies an internal logic chain consisting of four distinct stages from input to prediction: feature enhancing with shallow FFN neurons, feature transferring by shallow attention layers, feature predicting by arithmetic heads, and prediction enhancing among deep FFN neurons. Moreover, we identify the human-interpretable FFN neurons within both feature-enhancing and feature-predicting stages. These findings lead us to investigate the mechanism of LoRA, revealing that it enhances prediction probabilities by amplifying the coefficient scores of FFN neurons related to predictions. Finally, we apply our method in model pruning for arithmetic tasks and model editing for reducing gender bias. Code is on https://github.com/zepingyu0512/arithmetic-mechanism.

Kees Camfferman, Malte Max

Uit onderzoek van de iXBRL-jaarrekeningen van 67 in Nederland genoteerde ondernemingen over 2022 blijkt dat de nieuwe ESEF-vereisten met betrekking tot ‘block tagging’ van de toelichting door al deze ondernemingen zijn toegepast. Wel zijn er nog aanzienlijke verschillen tussen ondernemingen in de mate van detail waarin de toelichting wordt gemarkeerd. De vergelijkbaarheid – en dus de bruikbaarheid – van de informatie lijkt hierdoor vooralsnog beperkt. Regelgevers kunnen ondernemingen helpen door elementen uit de taxonomie meer eenduidig te definiëren.

Houssine Zine, Abderrahim El Adraoui, Delfim F. M. Torres

We propose a mathematical spatiotemporal epidemic SICA model with a control strategy. The spatial behavior is modeled by adding a diffusion term with the Laplace operator, which is justified and interpreted both mathematically and physically. By applying semigroup theory on the ordinary differential equations, we prove existence and uniqueness of the global positive spatiotemporal solution for our proposed system and some of its important characteristics. Some illustrative numerical simulations are carried out that motivate us to consider optimal control theory. A suitable optimal control problem is then posed and investigated. Using an effective method based on some properties within the weak topology, we prove existence of an optimal control and develop an appropriate set of necessary optimality conditions to find the optimal control pair that minimizes the density of infected individuals and the cost of the treatment program.

Anna Gold, Barbara Majoor

n/a

Rodrigo Rivera-Castro, Ivan Nazarov, Evgeny Burnaev

This research article suggests that there are significant benefits in exposing demand planners to forecasting methods using matrix completion techniques. This study aims to contribute to a better understanding of the field of forecasting with multivariate time series prediction by focusing on the dimension of large commercial data sets with hierarchies. This research highlights that there has neither been sufficient academic research in this sub-field nor dissemination among practitioners in the business sector. This study seeks to innovate by presenting a matrix completion method for short-term demand forecast of time series data on relevant commercial problems. Albeit computing intensive, this method outperforms the state of the art while remaining accessible to business users. The object of research is matrix completion for time series in a big data context within the industry. The subject of the research is forecasting product demand using techniques for multivariate hierarchical time series prediction that are both precise and accessible to non-technical business experts. Apart from a methodological innovation, this research seeks to introduce practitioners to novel methods for hierarchical multivariate time series prediction. The research outcome is of interest for organizations requiring precise forecasts yet lacking the appropriate human capital to develop them.

Jeff Johnson

The logarithmic number system (LNS) is arguably not broadly used due to exponential circuit overheads for summation tables relative to arithmetic precision. Methods to reduce this overhead have been proposed, yet still yield designs with high chip area and power requirements. Use remains limited to lower precision or high multiply/add ratio cases, while much of linear algebra (near 1:1 multiply/add ratio) does not qualify. We present a dual-base approximate logarithmic arithmetic comparable to floating point in use, yet unlike LNS it is easily fully pipelined, extendable to arbitrary precision with $O(n^2)$ overhead, and energy efficient at a 1:1 multiply/add ratio. Compared to float32 or float64 vector inner product with FMA, our design is respectively 2.3x and 4.6x more energy efficient in 7 nm CMOS. It depends on exp and log evaluation 5.4x and 3.2x more energy efficient, at 0.23x and 0.37x the chip area for equivalent accuracy versus standard hyperbolic CORDIC using shift-and-add and approximated ODE integration in the style of Revol and Yakoubsohn. This technique is a novel design alternative for low power, high precision hardened linear algebra in computer vision, graphics and machine learning applications.

Allam Maalla, Guang-Yu Wu, Shaojie Li

This paper discuss the basic thinking, methods and tools of software engineering, masters the financial business knowledge, the analysis and design theory and methods of financial information systems, and has the ability to analyze, design, implement and maintain financial information systems, and can be used in financial applications. IT companies such as development companies and financial information system providers engage in the analysis, design and implementation of software, or engage in the analysis, design, implementation, maintenance and management of financial information systems in the IT departments of various financial institutions such as banks, securities and insurance. A financial information-based composite software talent with a solid professional foundation, broad knowledge, and the ability to adapt to the future development of information technology. Specifically, it is reflected in four aspects: knowledge system, professional skills, project experience and comprehensive quality: Introduction On July 1, 2017, the General Office of the State Council issued the “Opinions on Strengthening the Service and Supervision of Market Subjects by Using Big Data”; on July 4, the State Council issued the “Guiding Opinions on Actively Promoting the “Internet+” Action” On September 5, the State Council issued the "Outline for the Promotion of Big Data Development." The intensive introduction of these heavy documents marks the official establishment of China's big data strategic deployment and top-level design. Benefiting from the rapid expansion of the big data market, the demand for related IT support has exploded. Among them, enterprises that provide big data infrastructure, big data software technology services, and industry big data content consulting services have brought unprecedented Customer group. IDC predicts that by 2020, the company's expenditure based on big data computing and analysis platform will exceed 500 billion US dollars, and the compound growth rate will reach 34.1% in the next 5 years; in the next 3 to 5 years, China needs 1.8 million data talents, but currently only About 300,000 people. At the same time, China's colleges and universities in cloud computing, data science and other majors are still in their infancy, and the talents cultivated each year are far from meeting the needs of the industry. Therefore, it is imperative to open a big data major and accelerate the cultivation of talents. The financial industry is the industry that relies most on data and is the easiest to realize data. In recent years, emerging financial institutions such as consumer loans and P2P are the products of the combination of big data technology and finance. At present, the demand for big data talents in finance is extremely strong in China. Only Internet finance is one year, and the growth rate is 3-5 times per year. It is generally believed that there will be a gap of 1 million talents in Internet finance, and the most lacking is big data risk control talents, including data mining and statistical modeling talents from primary to advanced. Core and Featured Courses Cloud Computing and Introduction to Big Data As an introductory course in the direction of this major, this course introduces students to the concepts, technologies and applications related to cloud computing and big data, and enables students to establish a preliminary understanding of the relevant knowledge, technology and development prospects of the profession, as a guide for the follow-up course. Distributed Computing Framework Foundation As the foundation and core technology course of this major, this course introduces students to the basic concepts, installation and configuration of the Hadoop distributed computing system, distributed programming model (Map/Reduce), distributed file system (HDFS), and related scheduling. , monitoring and maintenance tools enable students to build a basic understanding of distributed computing systems, master the primary distributed application design and implementation methods, and lay the theoretical and practical foundation for subsequent in-depth courses. Distributed Database Management and Development As a core technical course in this major, this course introduces students to the basic concepts of distributed databases, installation and configuration, management and maintenance, data access and development. The course focuses on NoSQL databases such as HBase, MongoDB, Redis, etc., and describes their use and development in a distributed environment. To enable students to establish a basic understanding of distributed databases, master the primary distributed database application system design and development methods, and lay the theoretical and practical foundation for the subsequent in-depth courses. Distributed Computing Framework Component Technology As a core advanced course in this major, this course introduces students to mainstream components on the Hadoop distributed computing platform, including Hive, Pig, Sqoop, Flume, Kafka, Zookeeper and more. Enable students to have a complete Hadoop ecosystem-based design and implementation of big data applications. Real-time Calculation and Memory Calculation As a core advanced course in this major, this course introduces students to high-performance distributed computing frameworks, including Storm and Spark, as a more powerful alternative to the Hadoop framework. Data Visualization Technology As an elective course in this major, this course introduces students to the basics of data visualization and the design and use of platforms and development tools, including Excel, Reporting Services, Chart.js, D3.js, Tableau, etc. Through this course, students will be able to present the results of big data processing in an efficient, flexible and friendly manner. Data Statistics and Analysis As a core advanced course in this major, this course introduces students to statistical analysis techniques based on Python and R. Including data file editing and finishing, basic statistical analysis, parameter estimation and hypothesis testing, non-parametric testing, analysis of variance, correlation analysis, regression analysis, cluster analysis, discriminant analysis, factor analysis, correspondence analysis, reliability analysis, survival Analysis, time series analysis, and the drawing of statistical graphs enable students to master the processing and analysis methods of typical industry business data. Knowledge System Mathematical basis: Including calculus, linear algebra, probability statistics, numerical analysis, etc. IT foundation: Including operating systems, networks, databases, software engineering, programming techniques, data structures and algorithms, etc. Knowledge base in the financial sector. Including international finance, marketing, insurance, securities investment, etc. Professional Skills Database system management and development: MySQL, MongoDB, Redis, HBase, etc. Big Data Application Development Language: Java as the core, supplemented by Python, Scala, R, etc. Construction, configuration, development and deployment of big data processing frameworks: Hadoop, Storm, Spark, etc. Use of data analysis and presentation tools: reporting tools, D3.js, etc. Project Experience Familiar with enterprise software project life cycle, development process, specification, etc. Understand and implement software quality requirements: performance, security, scalability, maintainability, reliability, etc. Understand the financial industry: industry background, business model, market characteristics, and how data and IT systems are used in the financial industry Comprehensive Quality Good professional basic qualities: document writing, presentation reporting, business communication, etc. Strong learning ability and study habits, has a certain degree of microinnovation, data awareness Course Settings Table The professional competence-course structure derivation process mainly includes two stages: “computation ability theme” and “capability-curriculum structure transformation”. The main process of the first phase of the "computational power theme" is as follows: Figure 1. The data analysis process of "ability topic calculation. Hadoop Big Data Integrated Experiment System This experimental system is designed to provide students with a complete set of Hadoop and its environment, design, development, monitoring, maintenance tools, software and services. With this experimental system, the experimental and training environment requirements of the core technology courses of this major can be met. This experimental system is divided into two major components: A virtual lab environment for students to learn big data. The environment is carried out by means of the aforementioned virtualized desktop teaching system, and the network administrator configures the big data learning virtual machine in advance for the students to use. The real environment for research or large-scale case presentations. This environment is carried out through several servers. Main Function A. Basic platform: The basic platform for big data storage and processing, which can realize the storage and management of massive data, support common components of platforms such as Hive, Impala, Pig, Spark, and Yarn, and provide support for data analysis services on the platform. These common components increase the ease of use of platform data, making data manipulation and data analysis easier to use, saving labor and reducing labor time. B. Data integration: support the unified storage of massive structured data, semi-structured data, and unstructured data, deepen the expansion of enterprise intelligence and service capabilities, and improve the decision-making level of enterprises. We can use enterprise-level data ETL tools or open source ETL tools. For example, Flume, Sqoop, Kafka, etc., integrate externally structured, semi-structured and unstructured data into big data platforms. Through this pla

Leonardo Munoz, Oscar Avila

Since technology has been involved in the business context, Business and Information Technology Alignment (BITA) has been one of the main concerns of IT and Business executives and directors due to its importance to overall company performance, especially today in the age of digital transformation. Several models and frameworks have been developed for BITA implementation and for measuring their level of success, each one with a different approach to this desired state. The BITA measurement is one of the main decision-making tools in the strategic domain of companies. In general, the classical-internal alignment is the most measured domain and the external environment evolution alignment is the least measured. This literature review aims to characterize and analyze current research on BITA measurement with a comprehensive view of the works published over the last 15 years to identify potential gaps and future areas of research in the field.

Sam Siewert

Tjerk Budding, Erwin Ormel

Met ingang van de begroting 2017 zijn de gemeentelijke verslaggevingsrichtlijnen ingrijpend vernieuwd. Een belangrijk deel van de vernieuwingen was erop gericht om te komen tot een betere vergelijkbaarheid en daarmee meer transparantie. Dit artikel gaat aan de hand van onderzoek met gebruik van meer dan 100 verantwoordingsstukken na in hoeverre deze doelen geslaagd lijken te zijn. Uit de resultaten komt naar voren dat de vernieuwingen wel formeel zijn doorgevoerd, maar dat de wijze waarop dat is gebeurd onvoldoende leidt tot een betere vergelijkbaarheid, en daarmee verbeterde transparantie.

Ali Enayat, Joel David Hamkins, Bartosz Wcisło

Ali Enayat had asked whether there is a nonstandard model of Peano arithmetic (PA) that can be represented as $\langle\mathbb{Q},\oplus,\otimes\rangle$, where $\oplus$ and $\otimes$ are continuous functions on the rationals $\mathbb{Q}$. We prove, affirmatively, that indeed every countable model of PA has such a continuous presentation on the rationals. More generally, we investigate the topological spaces that arise as such topological models of arithmetic. The reals $\mathbb{R}$, the reals in any finite dimension $\mathbb{R}^n$, the long line and the Cantor space do not, and neither does any Suslin line; many other spaces do; the status of the Baire space is open.

N. Karjanto, S. T. Yong

The Department of Applied Mathematics at the University of Nottingham Malaysia Campus has a responsibility for delivering mathematics modules for engineering students. Due to the significantly large number of students, two methods of teaching delivery--parallel teaching and block teaching--have been implemented. This article discusses some pros and cons between these two methods, particularly for the Foundation programme and the first year of the Undergraduate programme in Engineering. Whether parallel teaching or block teaching is implemented, feedback comments from the students indicate that some areas need to be paid attention to.

Ca Ochung, Benedict Mutuku, Jomo Kenyatta

Veronika Silinevicha, K. Kalinina

M. Rust, M. Nasir, Eric G. Meyer

Michel Planat

Phys and Math are two colleagues at the University of Sa{\c c}enbon (Crefan Kingdom), dialoguing about the remarkable efficiency of mathematics for physics. They talk about the notches on the Ishango bone, the various uses of psi in maths and physics, they arrive at dessins d'enfants, moonshine concepts, Rademacher sums and their significance in the quantum world. You should not miss their eccentric proposal of relating Bell's theorem to the Baby Monster group. Their hyperbolic polygons show a considerable singularity/cusp structure that our modern age of computers is able to capture. Henri Poincar{é} would have been happy to see it.