Search

We study a paper converting process where master rolls are transformed to paper sheets with the minimal paper loss, including trim loss, set up loss, and over-production loss. Two different types of cutters are involved in the converting process - one transversal knife and two transversal knife machines. To solve the problem, various types of paper sheets specified by material type, size, and required quantity are grouped into all possible cutting patterns using a heuristic method. Then, a MILP model is proposed to assign cutting patterns to each type of cutter while satisfying the machine capacity constraints. Applying the methodology to industrial problems, we found the average total loss of 4.6% for single-product group problems and 3.1% for two-product group problems, which respectively correspond to 25.8% and 41.9% improvement over a current method based on worker experiences used by the company. [ABSTRACT FROM AUTHOR]

Digitalization is a trendy issue in the business development world. It has experienced how vital the need for digitalization is during the Covid-19 pandemic process. Today's world needs digital transformation both to create a systematic data flow and to compete with competitors. Although the main reason behind digital transformation is to be able to compete because it helps businesses respond more effectively to the customer needs. Software product development differs from the conventional product development, so does the management of the project. In the last century, software development became a complex problem due to the change in customer needs. Thus, it requires different project management approach than traditional project management. This paper seeks to understand large-scale software development issue in terms of project management manners by reviewing literature. The paper will highlight the studies performed starting from 1992 until 2022. In doing so, the reader will observe the change in this issue over time. The papers are categorized into 5 groups according to their research questions. These groups are (1) project management and monitoring papers that focused on the governance of a project, (2) case study papers that they do not provide universal solutions but assess a specific case, (3) problem solving papers as they try to develop an algorithm to solve a problem during management processes, (4) explanatory/taxonomy papers which explains the terms and conditions of project management concept, and finally (5) literature review papers. To the best of authors' knowledge, the literature lacks a review study which provides an overview of project management of software product development concept that contains all project management frameworks in the literature. [ABSTRACT FROM AUTHOR]

Companion bots have the potential to alleviate loneliness and provide companionship. To accomplish this, the bot must generate appropriate conversations that are contextually relevant and interesting to the human companion. This research synthesizes 48 papers pertinent to companion bots. To understand and develop critical companion bot elements of the conversations, this paper explores prior literature to understand the human-interaction theories that have been leveraged. [ABSTRACT FROM AUTHOR]

A utility scale energy storage plant can be used on the North American (NA) grid for the storage of solar (PV) electricity and/or to provide ancillary services. To help bankers and their engineers determine whether hydrogen (H2) energy storage is ready for prime time2, this paper presents a H2 storage plant (HSP) levelized cost of storage (LCOS) financial algorithm for a model HSP. The LCOS algorithm is based on engineering economics and financial accounting principles. The algorithm only uses S. I. energy units for both the HSP's kinetic (electricity) energy and its potential energy (H2). It converts US$ values into € values. The paper discusses HSP technology focusing on the three phases of all HSP; charging the HSP by using a H2 electrolyzer (HE) to produce H2, storage of the PV electricity as H2 in a "H2 "Tank", and discharging the PV electricity from the HSP by using the stored H2 as the fuel for a fuel cell (FC) to regenerate the PV electricity. The LCOS algorithm uses "project accounting" to compute a separate "LCOS" for each HSP phase. The LCOS algorithm is presented in the paper's Excel HSP LCOS Financial Algorithm Workbook. Based on the paper's LCOS algorithm and on the HSP specifications (specs, metrics) complied for a base case "data set" for the model HSP, H2 energy storage is not yet ready for commercial development on the North American grid. Low round trip HSP efficiency (n) and a high HSP CapEx do not allow a HSP to operate commercially on the North American grid. [ABSTRACT FROM AUTHOR]

The emergence of edge computing allows the processing of data at the source of data. By combining edge computing with the cloud and the Internet of Things (IoT), data collection, processing, and storage can be managed effectively. This combined architecture has given hope for an automated surveillance system. This paper presents an edge–cloud – IoT-based Surveillance System that uses computer vision algorithms in the edge devices to process the video feed captured by cameras and store the required video backup in the cloud through IoT. This paper discusses the protocols used, task scheduling methods, metrics used to analyze the quality of service, and security and privacy issues. The dataset is collected from the SRMIST, Kattankulathur campus. The architecture is tested on the students' activities and is found efficient. [ABSTRACT FROM AUTHOR]

At the present time "chaotic nonlinear dynamical systems" is the most popular branch of the mathematical modeling. Many of real life phenomena are nonlinear. In the present paper we investigate the simulations of fractal figures by computer programming. Fractals are strictly dependence with the nonlinear phenomena. There was scientific view that every real phenomenon is regular or stable at the time of Isaac Newton. Later Poincare [2] observed many of the real phenomena are not regular i.e. they are "chaotic". At first time fractal figures observed on the computer by Benoit Mandelbrot with several programmers of at the company IBM in 1980. Later appears the set of Mandelbrot [4] which is fractal, the most important tool of the sets of Julia [1] and strictly depends on the irregular phenomenon. In this paper we learn the developing computer program to obtain fractal figures which applicable to many areas of natural phenomena. [ABSTRACT FROM AUTHOR]

Pharmaceutical industry is instrumental to the well-being of people around the world as it deals with the discovery, development, production and marketing of drugs and medications, which are one of life’s necessities. Recently, there is an emerging phenomenon that some hospitals in Vietnam must suppress their operation due to the deficiency of medical supply, and thus affecting patients who are in the need for treatment. One of the factors contributing to the situation is the incapability of providing these commodities on time from pharmaceutical companies, which results from their inaccurate forecasting. This paper investigates the application of machine learning algorithms in time series forecasting, specifically focusing on Support Vector Regression (SVR) to improve forecast accuracy. It evaluates the performance of two time series models, including ARIMA and Holt’s Trend, and three types of SVR, including Linear kernel, Polynomial kernel, and Radial Basic Function (RBF). The evaluation is based on the Mean Absolute Percentage Error (MAPE) metric, which measures the accuracy of the forecasts. The study then proposes hybrid models that combine SVR with traditional time series models via ensemble methods and compares their performance with the individual models. Finally, a case study of a pharmaceutical company is performed to assess the impacts of these methods. The results demonstrate that the hybrid ensemble method achieves lower MAPE values compared to the individual models. These findings offer promising insights and highlight the potential of the proposed hybrid models in improving forecast accuracy. [ABSTRACT FROM AUTHOR]

The traditional microblog recommendation algorithm has the problems of low efficiency and modest effect in the era of big data. In the aim of solving these issues, this paper proposed a mixed recommendation algorithm with user clustering. This paper first introduced the situation of microblog marketing industry. Then, this paper elaborates the user interest modeling process and detailed advertisement recommendation methods. Finally, this paper compared the mixed recommendation algorithm with the traditional classification algorithm and mixed recommendation algorithm without user clustering. The results show that the mixed recommendation algorithm with user clustering has good accuracy and recall rate in the microblog advertisements promotion. [ABSTRACT FROM AUTHOR]

With the advent of the internet there is vast increase in the storage of information. Almost all the information exists in digital form which reduces lots of the paper work and increases ease of storage. Searching relevant information in collection of documents is a tedious task. The solution comes in picture for this problem is automatic text summarization. In this paper, abstract summary generation of multiple documents for big data is proposed which will consider user input as topic. The proposed technique is designed using DBSCAN algorithm which works with Map Reduce framework for clustering and Hidden Markov Model for summarization. The summarization process is performed in three main stages and provides a modular implementation of multiple documents summarization. The pro-posed method follows preprocessing step in which documents are scanned with similarity and various machine learning technique are applied. The result of applying clustering enhances the summarizer system to collect exact words rather than copying redundant words. Topic based abstract summarization from big data is challenging task particularly when there are multiple documents with same or different content. Hadoop with its programming techniques can provide better ways of generating summary and it also enhances the complexity of summarization process using distributed computing. [ABSTRACT FROM AUTHOR]

With the number of Alzheimer’s patients reaching 5 million in 2014 according to the U.S. Center for Disease Control and Prevention, increasing emphasis has been placed on identifying and understanding its precursor condition, Mild Cognitive Impairment (MCI). MCI is characterized by subtle but abnormal cognitive decline and is challenging to detect without formal testing. Neuropsychologists use paper-and-pencil tests such as the Trail-Making Test (TMT) for diagnosis, and ongoing research places importance on high-granularity sketch data from digital TMTs. We present SmartStrokes, a digital TMT app designed to simulate the paper-and-pencil testing experience on a tablet and stylus. Our contribution frames the principles of digital sketch recognition and Human-Computer Interaction (HCI) into the existing neuropsychological test, outlining the creation of a pair of classification models that identify MCI on an individual segmented line basis. Such a per-line classification method which could provide localized sketching behavior indicative of MCI. We also present an interface for the digital TMT and a refinement of line segmentation algorithms from previous research to better distinguish between the actions that a participant takes when completing the exam. [ABSTRACT FROM AUTHOR]

Reproducibility of results is a central pillar of scientific work. In music information retrieval research, this is widely acknowledged and practiced by the community by re-implementing algorithms and re-validating machine learning experiments. In this paper, we argue for an increased need to also reproduce the results and findings of user studies, including qualitative work, especially since these often lay the foundations and serve as justification for choices taken in algorithmic design and optimization criteria. As an example, we attempt to reproduce the study by Kim et al. [1] presented in the RecSys (2020) paper “Do Channels Matter? Illuminating Interpersonal Influence on Music Recommendations”. By repeating this study on how interpersonal relationships can affect a user’s assessment of music recommendations on a new sample of n = 142 participants, we can largely confirm and support the validity of the original results. At the same time, we extend the analysis and also observe differences with regards to adoption rates between different channels as well as different factors that influences the adoption rate. From this specific reproducibility study, we conclude that potential cultural differences should be accounted for more explicitly in future studies and that systems development should be more explicitly connected to its intended target audience. [ABSTRACT FROM AUTHOR]

The paper focuses on understanding the perceived responsibility of marketing managers, and how Artificial Intelligence (AI) influences the acceptance of recommendations during the managerial decision process. AI’s impact is compared to the influence of human advice. Two experiments differentiate AI-based vs. human recommendations: The first study extends prior research on managerial decision making by showing that managers, who feel highly responsible for decisions, are less likely to accept recommendations in general. Interestingly, this effect differs between the source of recommendation. Managers tend to accept an algorithmic recommendation less when their perceived responsibility is high. In contrast, human recommendations do not differ by the degree of responsibility. The second study analyzes this moderation effect by including sense of ownership as mediator. Results show a significant indirect interaction between perceived responsibility and managers’ willingness to accept a recommendation through sense of ownership and source of recommendation. Overall, these findings contribute to understanding the psychological mechanism underlying the individual perception of responsibility and the acceptance of AI-derived recommendations. Thus, this paper offers valuable insights into improving AI acceptance. [ABSTRACT FROM AUTHOR]

This paper analyzes how the materiality of knowledge afforded by wearables is present within the practice of self-surveillance. Wearables, such as smartwatches, allow for easy tracing and storage of one’s activity history, but they require the wearer to engage in self-surveillance. While giving away one’s data to gain access to online services is motivated by efficiency and ‘free’ accessibility, the practice of self-surveillance in wearables is blurrier. In this paper, I situate the notion of desire as central to the practices of digitalized selfsurveillance. This paper follows the following line of inquiry: two case studies of recent self-tracking trends (Quantified Self Movement, Apple Watch) and one historical, media archeology case study of diary writing from the 19th century. The digital self-surveillance is linked to the notion of agency and control exercised by self-disembodiment. Disembodiment is briefly analyzed as establishing a disembodiment view of the self through a distant view of the past and presence. Such a view, ‘from outside to inside,’ is linked to the selfknowledge inherent in self-tracking across time. The body becomes a data object, at once a stable entity and an object-in-making. Yet there is a threat in perceiving data as an objective mirror of reality (and thus self) rather than a set of opaque algorithm-driven (‘cooked’) digital objects. [ABSTRACT FROM AUTHOR]

This research aims to improve the quality of life for visually impaired individuals by identifying obstacles in an indoor environment using a solution approach. The approach uses a Convolutional Neural Network (CNN) to extract features and detect objects from real-time video. A head-mounted image acquisition device detects objects and provides audio information to the visually impaired. The system processes live video streams frame-by-frame, processing each frame as separate images. The authors incorporate additional variables into a novel two-step iterative approach to solve complex computer vision problems. The proposed method improves computational efficiency and accuracy, opening up promising avenues for future research. The paper also presents a new mathematical formulation of curve and surface reconstitute algorithms by introducing auxiliary variables. Moving object detection is crucial for intelligent video monitoring systems, as it allows for accurate extraction of foreground objects. This paper aims to detect real moving objects from un-stationary backgrounds, limiting false negatives and achieving maximum application independence. The model and motion of the target objects are assumed to be unknown. Introduce a new mathematical formulation for curve and surface reconstruction algorithms, introducing auxiliary variables and minimizing energy. This approach transforms an implicit data constraint into an explicit convex reconstruction problem, simplifying it. The formulation also allows for more precise parameter settings, ensuring convergence to a minimum. And demonstrate the properties and results of this new auxiliary problem, mainly when the potential is a function of the distance to the closest feature point. [ABSTRACT FROM AUTHOR]

Artificial Intelligence (AI) allows computer programs to learn from experience through iterative processing and algorithmic training and represents human intelligence. Now, AI has become a hot topic and important debates have risen like, how it might affect the job market and what is the end of global civilization. Many people will lose their jobs due AI. On the other hand, few technology lovers think that it will lead to the creation of lots of diversified and interesting jobs. Many experts think that it will have a big impact on the workplace and all aspects of human life in the near future. In reality, AI can make jobs more creative, lucrative, and flexible, and will lead definitely to a more creative and skilled economy. However, history says, that advanced technologies are usually starting to take away more jobs than they create, and unfortunately, this trend will continue. Nowadays, there have been many ideas for how to solve this problem, and there should be updated education and skill develop program with new schooling system and that need to be introduced very soon. As there isn't enough work for everyone in the world, so the whole lifestyle and education system needs to be changed and to reorganize every aspect to make total human resource development system more purposeful. It is an analytical paper to analyze the potentials and challenges of AI along with the consequences of global future and way forward. [ABSTRACT FROM AUTHOR]

Copyright of ESAIM: Proceedings & Surveys is the property of EDP Sciences and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Most of the mobile robot path planning is estimated to reach its predetermined aim through the shortest path and avoiding the obstacles. This paper is a survey on path planning algorithms of various current research and existing system of Unmanned Ground Vehicles (UGV) where their challenging issues to be intelligent autonomous robot. The focuses are some short reviews on individual papers for UGV in the known environment. Methods and algorithms in path planning for the autonomous robot had been discussed. From the reviews, we obtained that the algorithms proposed are appropriate for some cases such as single or multiple obstacles, static or movement obstacle and optimal shortest path. This paper also describes some pros and cons for every reviewed paper toward algorithms improvement for further work. [ABSTRACT FROM AUTHOR]

Along with digitalization and the emergence of data ecosystems, organizations face an intransparent field for innovation. Data-driven business models arise in different social and industrial sectors, new sensors and devices break down the barriers for disruptive ideas and digital transforming established solutions. This paper aims at bringing light into the darkness and give guidance to the reader which topics in the field of data economy are hot or not in relation to creating innovation. Therefore, the paper uses a text mining supported systematic literature review to automatize the extraction and analysis of beneficial insights for both scientists as well as practitioners that would not be possible by a manual literature review. By doing so, we were able to analyze 860 scientific publications resulting in an overview of the research field of data economy and innovation. Nine clusters and their key topics are identified, analyzed as well as visualized. [ABSTRACT FROM AUTHOR]

This paper proposes a model for measuring video content visual activity based on a computer vision algorithm. Several authors have implied to certain trends that films and other video content have been imposing during the previous decades onwards. Firstly, the duration of shots is decreasing, the scenes are darker and there are more visual activities in films than before. To indicate and demonstrate the previous three trends, other simple models have been used by other authors. In this paper one background subtraction algorithm has been used to measure and verify increased visual activity in video content. The algorithm has been used to process the video clips of the MTV Music Awards winners, which were the winners in the category of Video of the Year in the last 30 years. [ABSTRACT FROM AUTHOR]

This paper presents a method for generating property graphs from OpenStreetMap data as a precursor to track generating methods for cycling sports. The results indicate that OpenStreetMap geographical data can be represented on a property graph. This is beneficial, and needed for use of computational intelligence path generation algorithms. The paper is concluded by presenting a sample property graph representing roads and intersections as nodes and relationships based on OpenStreetMap geographical and EU-DEM elevation data. [ABSTRACT FROM AUTHOR]

In this paper, we raise practical situations in profiling based power analysis when profiling and target devices are quite different at several levels. "Crossed devices" are newly termed, including homogeneous and heterogeneous devices, which have not been carefully investigated. We identify such device variations and take a further step towards leveraging the deep learning based power analysis. Traditional template attacks and straight-forward deep learning based power analysis will fail, when the gap across devices is significantly enlarged. In this paper, we propose a noval frequency and learning based power analysis machanism, which is able to explore new attacking power of deep learning and address challenges caused by device variations. For the first time, power traces collected from our own PIC devices can be utilized to successfully attack the public dataset in DPAContest v4 which is based on a totally different AVR microcontroller. [ABSTRACT FROM AUTHOR]

Today, the study of the distribution of emissions in the atmosphere is quite an urgent task, because harmful substances, getting into the surrounding atmosphere, cause harm to the health of people, flora and fauna. In this paper, the results of simulation of the distribution of emissions from the Ekibastuz TPP (GRES 1) are presented. For simulations the model of turbulence was chosen. Simulations were performed by using the ANSYS 16.0 software package. Previously, the model and numerical algorithm were tested using a model problem and the results were compared with the experimental data. [ABSTRACT FROM AUTHOR]

With recent advances in artificial intelligence, machine learning has been identified as particularly useful for organizations seeking to create value from data resources. However, this usefulness presupposes the existence of data with sufficient structure and quality to train machine learning models. Thus, in this study, we address the research question: How does data resourcing contribute to enabling AI opportunities? We address this question by investigating an example of data resourcing in a master’s level artificial intelligence (AI) course at UmeÃ¥ University, Sweden. This is an empirical, case-based educational setting where students engaged in data resourcing using a no-code AI platform. Our results involve provide insights regarding constructs associated with two phases of data resourcing: data work practices, and data interpretation. Collectively, these constructs provide a framework for data resourcing that is the main contribution of this paper, together with insights into the benefits of using no-code AI tools in educational settings. Our framework reveals the nature of data resourcing as a creative process where iterative algorithmic mediation, data sensemaking and contextualization enables AI opportunities. With recent advances in artificial intelligence, machine learning has been identified as particularly useful for organizations seeking to create value from data resources. However, this usefulness presupposes the existence of data with sufficient structure and quality to train machine learning models. Thus, in this study, we address the research question: How does data resourcing contribute to enabling AI opportunities? We address this question by investigating an example of data resourcing in a master’s level artificial intelligence (AI) course at Umeå University, Sweden. This is an empirical, case-based educational setting where students engaged in data resourcing using a no-code AI platform. Our results involve provide insights regarding constructs associated with two phases of data resourcing: data work practices, and data interpretation. Collectively, these constructs provide a framework for data resourcing that is the main contribution of this paper, together with insights into the benefits of using no-code AI tools in educational settings. Our framework reveals the nature of data resourcing as a creative process where iterative algorithmic mediation, data sensemaking and contextualization enables AI opportunities. [ABSTRACT FROM AUTHOR]

Monte Carlo rendering algorithms can generate high-quality images; however they need to sample many rays per pixel and thus are computationally expensive. In this paper, we present a method to speed up Monte Carlo rendering by significantly reducing the number of pixels that we need to sample rays for. Specifically, we develop a neural future frame synthesis method that quickly predicts future frames from frames that have already been rendered. In each future frame, there are pixels that cannot be predicted correctly from previous frames in challenging scenarios, such as quick camera motion, object motion, and large occlusion. Therefore, our method estimates a mask together with each future frame that indicates the subset of pixels that need ray samples to correct the prediction results. To train and evaluate our neural future frame synthesis method, we develop a large ray-tracing animation dataset. Our experiments show that our method can significantly reduce the number of pixels that we need to render while maintaining high rendering quality. [ABSTRACT FROM AUTHOR]

This article presents a customized approach for training a supervised learning neural network with the adaptive moment estimation algorithm, to classify the number of frequency hopping networks in an operational area. The algorithm was constructed based on data experimentally collected from a real-time spectrum analyzer for military very high frequency hopping networks. The impact of some training parameters on classification efficiency is briefly discussed while the obtained accuracy was above 97% for both test and validation data in all training variations. With this promising result, the proposed algorithm has the potential to be utilized in developing operational systems capable of real-time signal reconnaissance for military frequency hopping radio networks. [ABSTRACT FROM AUTHOR]

Computer graphics is the Visual Presentation of diagram, photographs that we draw on a surface with the help of Computer. We can transform the data or information in visual form and can represent that on any display device. Numerous drawing algorithms have been used in computer graphics for rasterizing the screen. This paper deals with a new way to reduce the error in the accessible Midpoint Circle Drawing Algorithm. In this approach, we have measured two dissimilar values of h as (1 and 0.1) in the first quadrant of a circle. For h=1, every one of the outcomes of Midpoint Circle Drawing Algorithm have been confirmed and for different values of h it is seen that as the value of h diminishes, the quantity of cycle increments yet the errors between the points created and the first circle points diminishes and the other way around. Finally by doing symmetry a whole circle is been drawn. [ABSTRACT FROM AUTHOR]

Computer graphics is the illustration and representation of image data by a computer employing several technologies to construct and operate images. Visualization, graphical user interface, graphical display and presentation, manipulation of mathematical objects and rendering, motion capture, etc. are some of the important subjects involved in computer graphics. A number of drawing algorithms have been developed so far. Their underlying principle is to capacitate visually presentable images in a minimal time. This is ultimately accomplished by lessening the calculations to a minimum. Ellipse is one of the forms of the Circle, when we stretch the circle then the resulted figure is known as Ellipse. In this paper we have used a new method to generalize the Midpoint Ellipse Drawing Algorithm which reduces the error in the existing one. In this method, we have considered different values of h as (1, 0.1). MPEDA is verified for the all results when h=1, and the values of h is decreased for other values of h, in this the error between points are generated when the number of iterations increased. Similarly the Ellipse points decreases and vice versa. [ABSTRACT FROM AUTHOR]

Within the context of high precision robotics fields such as surgery, machinery requires the utmost accuracy. This paper introduces the method of proportional gyroscopic movement, applied through iterative testing with Lego Mindstorms EV3 technology. Consequences of unstable robotics on an industrial level can be as severe as death of a patient. To counteract the influence of external motion upon the linear movement of a robot, the gyroscopic algorithm (“gmove”) being tested was applied to an educational Lego EV3 rover robot to perform along a laser straight reference line. Across all trials, gyroscopically oriented movement demonstrated superiority in both precision and consistency, bringing to light a novel method of precision movement in the robotic industry. [ABSTRACT FROM AUTHOR]

The paper deals with humanoid robots and its interaction inside open areas. The major part of the paper presents the space scanning algorithm which corrects movement imperfections of humanoid robot during exploration of space. Algorithm utilizes visual confirmation retrieved from data collected by robot, what allows to make robot movement much more precise in specific space. [ABSTRACT FROM AUTHOR]

There is much interest in developing solutions for protecting data privacy in recent years, and many privacy models and data sanitization methods have been proposed. However, relatively little has been done to understand how existing data analysis techniques may be adapted to work with sanitized data. In this paper we report a study on learning decision trees from anonymized data. We sanitize data using the Mondrian algorithm to satisfy k-anonymity and adapt the ID3 algorithm to learn decision trees from sanitized data. Our preliminary experiments show that accurate decision trees can be learnt from anonymized data, and degradation of classification accuracy is no more than 2% with typical settings. [ABSTRACT FROM AUTHOR]

This paper is conducted to develop a system that can aid universities in tracking students' academic progress by accurately determining dropout rates using machine learning techniques. This has been done by collecting and organizing data into machine readable forms, using predictive methods that determines significant patterns and trends, constructing different models involving machine learning techniques: Random Forest, Logistic Regression, and Neural Network, for forecasting and predictions, and finally interpreting the statistical results of the model. After the evaluation, it has been concluded that the Random Forest achieved the highest accuracy of 95.00%, precision of 97.78%, and F1 score of 96.70%. Despite these results, the researchers recommend developing a more interesting and engaging prototype GUI and utilizing a more balanced and diverse dataset that will enable higher accuracy and deeper understanding of the results. [ABSTRACT FROM AUTHOR]

The benefits and application potentials of artificial intelligence in controlling are widely known. Contrary to expectations, an in-depth interview study of Austrian industrial companies shows that the use of AI in controlling is only just beginning. In order to make controllers fit in this area, in-depth knowledge channels are required. In addition, several other aspects need to be considered for implementation. This paper presents the results of a qualitative pilot study from Austria on the use of artificial intelligence in the controlling departments of Austrian industrial companies. [ABSTRACT FROM AUTHOR]

With increasing number of applications of Unmanned Aerial Vehicles (UAVs) in civilian sectors, enthusiasts are finding ways to build their own UAV which can be used for jobs that are too dull, dirty or dangerous for humans. Design of UAVs may vary according to their respective mission and payload requirements. The algorithm asks for the specific requirements of the user such as payload weight & volumetric dimensions, maximum speed at which UAV is supposed to fly etc. The process includes aerodynamic analysis of numerous airfoils and selecting the optimum one, calculating wing-span required with the selected airfoil cross section to attain required lift, reducing drag, maintaining stability of aircraft etc. This process is iterative and can take lot of time. Here arose a need to make this process automatic with the help of computer by developing an algorithm which can save lot of time for the people who have to work from scratch in order to make an UAV. In this research paper an effort is made to establish a computer-based algorithm for design and development of a flying wing UAV using MATLAB. [ABSTRACT FROM AUTHOR]

The aim of the research presented in this paper is to propose a conceptual model and architecture of a service-oriented scalable framework, ensuring the implementation and verification of methods and algorithms for the integration, management, analysis, and visualization of biomedical data and the implementation of scientific research for the needs of precision medicine. The system architecture for big biomedical data analytics and discovering useful knowledge from data consists of the following components: biomedical data sources, data storage, data integration and preprocessing, real-time data flow, stream processing, analytical data storage, data modeling and analysis, and results visualization. A feed-forward artificial neural network is designed for data analysis, and during the training process, the input data is divided into training data and test data. The training error and its distribution over the weights of the neurons in the network are determined. A reduced set of statistical records related to cardiovascular disease analysis has been used as experimental data. The original database contains 76 attributes, and 14 of them have been used for the study. In addition, the data is split in a ratio of 0.8 to 0.2. The first 80% of the data was used to train the neural network and the remaining 20% to test the trained network. The calculated accuracy increases with increasing epochs and is higher for the training data and lower for the validation test data. Thus, the trained model can be saved, and loaded on another system, as well as available for review of the weight values. The trained model is applied in the system to calculate new input parameters that were not used either in training or validation. [ABSTRACT FROM AUTHOR]

In the era of technology, information or data is an important factor to do any kind of task. Recommendation system has been introduced in the tourism sector where the system can be used to provide the user with personalized recommendation of the places to visit. Sultanate of Oman has got various tourists attractions and it is one of the safest tourist destinations in the world. The purpose of this paper is to design and develop a recommendation system for tourism sector in Oman which will recommend to the users based on their searched destinations using hybrid recommendation system that is combination of- content-based and collaborative filtering algorithm. To train and then to provide meaningful recommendations, dataset has been developed by implementing web scrapping method from one of the popular websites ‘tripadvisor’ which consists of the description of all the tourism places in Oman namely destination names, type of the destination, average rating of the places, the number of people rated the place and image of the destination. The recommendation system is integrated with a web application that acts as a web interface. The system has been developed using python programming language while the web application was developed using Django framework along with HTML and CSS. To evaluate the efficiency and performance of the developed system, it was compared with three other algorithms namely Radius Neighbors classifier, Linear Regressor and Logistic Regressor. The same dataset was provided to all the three algorithms where 70% of the data is used as training data and the rest of the 30% of the data is used for testing. MSE (Mean Square Error) and RMSE (Root Mean Square Error) are calculated for all of the mentioned algorithms, and it was found the developed system performed better that most of the algorithms with less error. [ABSTRACT FROM AUTHOR]

The construct and instrument development process relies significantly on human judgment in the initial stages of the process, specifically in developing construct definition statements, and in developing measurement instruments with high content validity. Natural language processing (NLP) techniques can be used to support human judgment and improve the quality of constructs and instruments employed in research. This paper describes the use of NLP techniques in the construct and instrument development process and presents illustrative results from the use of those techniques. We develop an NLP-based algorithm and illustrate its use to assess measurement instruments. The data used to train the algorithm was 37 years of constructs published in premier IS journals during the period of 1980-2016. The empirical illustrations support our premise that the use of NLP techniques can improve the rigor of the process and improve the quality of constructs and instruments employed in research. [ABSTRACT FROM AUTHOR]

The dissemination of fake news intensified during the COVID-19 pandemic. Thus, in this paper we propose a process-based model grounded on the diffusion of innovations theory to investigate how fake news was shared via social media. Accordingly, data was categorized according to the taxonomy of process, namely: input, output, players and activities. We found that individual decision-making traits, fake news perceived features and communication channel attributes are determinants of a personal attitude vis-à-vis a received fake news. The attitude then leads to the decision of whether or not to believe in the fake news, while confirmation conveys to the spread of same. In this way, social media, mediated by the power of algorithms and echo chambers, increase the chance of spreading misinformation. [ABSTRACT FROM AUTHOR]

Recommender systems are used with frequency so that content/items are offered in a personalized way for each user, and it is important that these algorithms can accurately recommend content/items to these users (good predictive performance), as well as have a satisfactory computational performance - so that it is not necessary to use too many computational resources. Thus, this paper aims to evaluate some recommendation algorithms that use the memory-based Collaborative Filtering (CF) technique and to evaluate the influence of similarity metrics on the performance of these algorithms. Both algorithms and metrics are available in the scikit-surprise library. Two public databases were used: MovieLens 100k and MovieLens 1M. After the experiments, it was observed that the choice of similarity metric might influence the predictive performance and the prediction and training time of the algorithms. The MSD metric was the one that stood out in influencing, in a positive way, these results. It was also noticed that the database could influence both the predictive performance of the algorithms, as well as the RAM consumption. [ABSTRACT FROM AUTHOR]

Machine learning techniques inspired by the structure and function of the human brain are known as artificial neural networks (ANNs). Their popularity has grown in recent years due to their ability to learn and improve via training, allowing them to be used in a variety of applications. The ANN approach has the appealing advantage of providing an accurate representation of the process input/output data and due to this reason, their application in the extraction metallurgy field have increased over the years. This systematic review aims to investigate the application and efficacy of ANN for facilitating and upgrading mineralogical data processing in the extraction metallurgical field. In this study, 112 research papers, published over years, that discuss ANN application were retrieved and 49 research papers were systematically reviewed. Research trends, ANN algorithms models, and evaluation methods primarily discussed in the 49 papers were discussed. From the review, it was discovered that the main justification to the increase application of ANN in the extraction metallurgy field is due to the ore grade decline which makes extraction processes complex demanding a need for a predicting and optimising tool. The frequently used ANN algorithm was found to Levenberg-Marquardt algorithm (LMA) which was due to its ability to process a lot of data in a short period of time. It was further concluded that the applicability of ANN in metallurgical field diverse from processes optimisation using operational to equipment parameters with least error and high efficiency. [ABSTRACT FROM AUTHOR]

On-demand platforms operating on a gig-contractor model rely on self-scheduling gig workers to meet realtime demand, while maintaining desired levels of productivity and service quality. Such platforms usually allocate gig workers to tasks on a first-come-first-serve basis, which is often inefficient. This paper seeks to investigate whether allocation of workers to tasks based on prior experience and task characteristics could act as an alternative to the FCFS method. Since on-demand platforms frequently expand to new geographies where workers might not have prior experience, it is unlikely that allocating tasks to workers based on experience alone would significantly improve performance. We investigate whether recent experience can positively impact gig workers' task performance, and thus be utilized as a ranking parameter in task allocations. Utilizing data from an on-demand grocery platform, we first developed an econometric model to analyze gig workers' task performance based on their same-day experience, while accounting for sample selection and endogeneity. We find that as same-day experience increases, task productivity and service quality improves, even after controlling for worker and task characteristics. We developed two task allocation algorithms to allocate tasks based on workers' rank, and task characteristics. We calculate predicted improvements in performance from the new allocations. [ABSTRACT FROM AUTHOR]

Individuals increasingly seek algorithmic advice to optimize their decision making. This study aims to investigate the effects of receiving algorithmic advice on individuals’ attribution of responsibility for their achievements. The study is based on an experiment with a 2 x 5 design of two dimensions: achievement (success vs. failure) and advice (no advice; human-based advice with high and low expertise; and algorithmic advice with high and low accuracy). The findings from a pilot study suggest that the experimental design is largely appropriate, given that we found answers to our hypotheses. This short paper provides valuable insights for future research on the attribution of responsibility for success and failure when receiving algorithmic advice. [ABSTRACT FROM AUTHOR]

Modern-day firms face the predicament of blending the comparative advantages of their two core resources: machines and humans. When forecasting demand (e.g. for a product), extant literature documents that always permitting (or prohibiting) human revision of a machine forecast is beneficial if the humans’ private information role is larger (or smaller) than that of machine-accessible public information. We propose and test a complementary framework that shifts the focus to the regulation of each human revision; and in doing so, adjusts for human vulnerability to systematic biases. We collaborate with a European retailer to compile a large dataset (∼ 1.1 mn transactions) on machine-led demand forecasts. Humans revise nearly 38% of these forecasts, but revisions do not always yield an improved final forecast. Compared to the always permit or prohibit strategies, the ideal regulation of each of these revisions could reduce the absolute deviation of the final forecast from realized sales in close to 50% of instances. Any regulation to obtain the best of the machine and human-revised forecasts requires an ability to predict revision quality. In addition to private information indicators, we show that indicators of systematic bias in human judgement (e.g., indicators of cognitive load) can improve the prediction of revision quality. In an out-of-sample analysis, our revision-level regulation approach picks the best of available forecasts in 14% more instances, compared to an always-permit or prohibit strategy at the product-store level. Our paper provides practitioners with a novel approach to combining the machine-human output to improve demand forecasting performance. [ABSTRACT FROM AUTHOR]

Contemporary education promotes the usage of digital games to support learning programming. Such games are not designed for entertainment, but specifically for educational purposes. Students start from a very early age to interact with computers through games and consequently, react positively on the educational digital games. When used for education, digital games have the potential to motivate students towards active participation and interaction. In addition, educational digital games make complex and abstract topics like programming more friendly. Games that support learning programming enable students to learn programming concepts using visual interfaces and interesting environments. In these games, younger students are usually expected to move the main character around the given path and perform a series of tasks (e.g. collect objects, skip the obstacles). Instead in a textual editor, students are coding through drag and drop interactions, and create programs that include basic programming concepts like sequence, variables, loops, and conditions. Furthermore, digital games are a good for visualizing the execution of algorithms that are often hard to understand. To make the game more interesting, the main character is often presented as an animated robot, animal or a cartoon character familiar to the students. The paper explores the possibilities of using existing digital games in learning basic programming skills, highlighting which programming concepts the game supports and for which it is recommended and appropriate to be used. This is the first step in the design of an educational game that is intended to develop within the “Coding4Girls” and "Digital games” projects. Future activities and plans for both projects are presented in the paper. [ABSTRACT FROM AUTHOR]

In this paper, sowing rate optimization of grassland drills consisted in finding an optimum combination between the seed flow of the seed metering device and the drill working speed, in order to achieve the sowing rates required by agro-technique for herbs seeds. It was developed a mathematical model that describes the seed flow variation depending on the working speed of the grassland drill and the active length of the groove cylinders. The experimental research was conducted using a grassland drill with eight metering devices and an automated installation, which simulates the working speed of the drill, collects and weighs seeds from each individual metering device. By using MathCad 14.0, it have been drawn the variation diagram of seed flow depending on active length of the groove and working speed. The comparative study of the variation diagrams based on experimental data and the variation diagrams obtained from mathematical modelling highlighted that the downward trend of the seed flow was observed in 66,67% of the results. Based on the mathematical model presented in this paper, it might be developed an algorithm enabling for fast obtain the optimal combination between the adjustment parameters, in order to apply a desired sowing rate. [ABSTRACT FROM AUTHOR]

Artificial intelligence (AI) is a field of human interest since the 1950s. In recent years, it has gained widespread attention due to the needed underlying technologies now available. AI technology is based on neuronal networks that require large amounts of data and fast computation to process learning algorithms, which help organizations greatly in their decision-making processes. This paper explores AI applications in the automotive industry of the future with respect to the automotive value chain. The applications of AI are analyzed for all the subprocesses in the automotive value chain, which include: Research and Development, Procurement, Logistics, Operations, Marketing, Sales, and Connected Customer. The AI applications for each sector are investigated in terms of risk and impact on businesses in the automotive sector. This helps to identify, from an organizational perspective, the most applicable areas of AI technology in the future automotive industry. The recommendations made at the conclusion of this paper include explanations of how the AI application has to be made as part of the automotive value chain and the expected benefits of implementation. In addition, the findings are summarized in a table in the recommendations section, which shows the possible application areas of artificial intelligence in the automotive industry value chain along with the benefits and risks of the use-cases. [ABSTRACT FROM AUTHOR]

Nowadays in India, due to the increase in the number of accidents vehicles are being automated using a variety of Computer vision algorithms. This paper focuses on the detection of tail signals of cars under different illumination circumstances. This system is implemented using FAST and SIFT algorithm which helps to extract features from the images. The obtained features were optimized by using K-Means Clustering Algorithm. This huge feature vector is converted into 8 clusters. These optimized feature vectors were trained on five different classifiers as Decision Tree, SVM(RBF), Random Forest, and Voting Classifier. The trained data set used in this algorithm contains around 9052 images. The obtained accuracy results of different classifiers are as follows, Decision tress has 76.36%, SVM(RBF) has 77.91%, Random Forest has 87.52%, KNN has 89.72%, the Voting classifier has 86.52%. It is observed that KNN gives the highest accuracy among the five used classifiers. This has, to the best of the authors' knowledge, not been presented in literature before. [ABSTRACT FROM AUTHOR]