Abstract: The paper solves some problems belonging to the field of recognition of asocial behaviour in online space. Nowadays, it seems to be an important issue when we must question our way of dealing with the pandemic crisis. Social network users must deal with such unhealthy phenomena in online space as toxic comments and toxic troll authors that prevent constructive communication and knowledge sharing through the web space. We have proposed a new multimodal approach to social network analysis, which combines two methods, the first one to recognize toxic posts using machine learning and the second one to identify toxic authors in online space using sentiment analysis. The recurrent neural network was trained with different numbers of neurons in the hidden layers using three different types of hidden layers and optimizers along with various learning rates. Finally, the paper provides detailed results of extended experiments with deep learning models for recognition of toxic reviews, where a model generated by a combined deep learning architecture achieved accuracy over 0.9, and results of our novel approach to the detection of toxic troll reviewers achieving accuracy of 0.95. Our approach to troll recognition is based on a comparison of the sentiment related to the authors’ posts to sentiment related to all comments of an online discussion.

Keywords: web mining; data analysis; recognition of toxic posts; detection of toxic reviewers; trolling; deep learning; sentiment analysis; short texts processing

Abstract: In this study we show that sometimes it is worth to consider a non-traditional regular grid in diagrams and proofs by diagrams instead of the traditional rectangular arrangement. We refer to the hexagonal and triangular grids as non-traditional regular grids. Particularly, we show that not only triangular numbers can be arranged in triangular shape, but also, it is natural to represent the square numbers via triangles on the triangular grid. On the other hand, we show that hexagonal arrangement is useful to represent binomial coefficients, and also to solve jug puzzles. The triangular and hexagonal grids are dual of each other and share many properties including symmetric properties. Based on our study, we propose to consider them more frequently in education and diagrammatic applications as well. This would be fruitful also for critical thinking, design and problem solving in mathematics, engineering and informatics including artificial and computational intelligence.

Keywords: Computer-aided Education; Non-traditional Diagrams; Geometric Proof by Diagrams; Human-Computer Interaction; Data Visualisation; Number of Shortest Paths; Figural numbers

Abstract: During laparoscopic surgery, surgeons carry out surgical procedures with the assistance of a video camera and several surgical instruments. A critical skill for a successful operation is to develop a sense of the applied forces on the tissue using laparoscopic tools. A novel technique to measure the forces applied by the needle driver during artificial bowel suturing tests is proposed in this paper using strain gauges (SGs). A method to aggregate the responses of the SGs on a cylindrical shaft is also proposed. Another innovation involved in this study is a force feedback display system that utilizes Augmented Reality (AR) glasses that provide a real-time visual representation of the force to the surgeon without obscuring the field of view. An electromechanical calibration system to calibrate the force measurements for the needle driver and acquire force data using LabVIEW is also presented. This study is part of a collaborative research partnership between the Department of Electrical and Computer Engineering and the Department of Surgery of the Homer Stryker M.D. School of Medicine, at WMU.

Keywords: Laparoscopic Surgery Skill Test; Bowel Suturing Force Measurement; Needle Driver Force Measurements Using Strain Gauges; AR Glasses for Real Time Feedback

Abstract: The goal of this paper is to investigate the details of final-energy generation technologies for industrial energy systems. Currently, sustainable development, environmental aspects and economy are of paramount importance and these technologies play a very significant role in all of them. Moreover, the unpredictability of political, ecological and economic conditions poses a significant uncertainty for companies dealing with these products. To address this problem, a technology road-mapping framework was developed to facilitate the understanding of the environmental and technological maturity effects on these products, from the supplier companies’ point of view. In the frame of this investigation, the macro- and micro-environments were examined, then summarized, with the resulting opportunities and threats for the manufacturing firms. Based on the concluding roadmap, mid- and long-term trends can be identified, which are then available to support company-specific research.

Keywords: technology roadmapping; energy; final energy generator; TRM

Abstract: This paper describes the development of a unified cloud service, for protecting and monitoring corporate computer networks and SOHO-class networks, with intelligent mobile software and hardware clients, based on the Raspberry Pi type microcomputer. It is planned to develop an intelligent algorithm, that performs automatic decision-making and provides recommendations, when threats are detected in the network, to finalize software and hardware, taking into account the requirements of mobility and integrability, within the cloud service. The intelligent data processing algorithm, implemented on these devices, will be based on the developed linguistic processor and the procedure of automatic assessment of network threats. Implementation of this idea consists in the development of a web service with a replenished database of threats, incidents at the nodes of a computer network and standard solutions, a system for quantitative and qualitative risk assessment, as well as, in the subsequent integration of the described intelligent mobile software and hardware device into this web application. Thus, the developed cloud service, for protecting and monitoring computer networks, will be a centralized toolkit for the joint fight against network attacks, identifying vulnerabilities in the configuration of enterprises’ networks, for using and building up a database of investigated incidents and solutions, taking into account the permissible degree of data disclosure. This work presents a developed sequential plan for the implementation of this task. Attempts have been made to substantiate, theoretically, the feasibility of implementing the described task. The authors describe the relevant tools that are currently available.

Keywords: corporate networks; computer networks security; cloud service; intelligent system; single-board computer; Raspberry Pi

Abstract: Conformance checking is a class of process mining techniques, which contrasts the modeled behaviors with the observed behaviors of a process, to detect, locate and explain the deviations between them. Even though deviations can occur anytime and in any perspectives, currently there is no such conformance checking technique available, which is able to take into account other perspectives than the control-flow perspective of the investigated process when computing the conformance statistics on running, incomplete cases. In this paper, a multi-perspective online conformance checking technique is introduced, which aims to confront the modeled behaviors in form of a Data Petri net process model with a stream of events as the observed behaviors. For the conformance checking, two existing techniques were merged: a prefix-alignment-based technique, which is able to compute conformance statistics for incomplete process executions by applying incremental heuristic search, and an alignment-based multi-perspective conformance checking technique, which is able to compute conformance statistics for complete process instances while focusing on multiple perspectives.

Keywords: process mining; conformance checking; multi-perspective; prefix-alignment; incremental heuristic search; event stream; Data Petri net

Abstract: Torsional torque is considered as one of the loads that the crankshaft is constantly transferring. This force causes torsional stress (torsional stress due to torsion) in the fixed crankshaft bearings and shear stress in the movable bearings. Due to the effect of different parameters in the balancer design, e.g. inertial forces, the mass of weights, off-center, and distance from the axis of the crankshaft the need for extensive research in this field is sensed. The research investigates the effect of the balancer on reducing engine vibration. The main objective was to study the kinetics of the engine using Adams engine software by application experimental data. And finally comparing test results with software results to confirm created simulation. By analyzing the output signal, the root mean square (RMS) value of experimental data, the balancer reduced the amount of RMS value of engine vibration at 750 rpm in no-load mode by 40.46%, and 12.73% at the rotational speed of 2400 rpm. Accordingly, using the balancer with no-load was more effective at lower speeds. In the case of no-load mode, the use of a balancer reduced vibrations by 28.83%. It was found that in full load condition, the RMS of the engine vibrations increased with increasing engine speed. Experimental and simulated by software results were highly consistent with experimental results, which confirms the validity of the created simulation.

Keywords: Computational Analysis; Balancer; Vibration; Diesel Engine; Torsional Torque; Crankshaft

Abstract: To increase the reliability of energy distribution, it is necessary to shorten the downtime. It is very important to determine the location of a fault as accurately and as quickly, as possible. Than the maintenance service team can rectify the fault as soon as possible. Determining the distance of the fault (from the relevant power station) with the greatest possible accuracy, in the event of a short circuit on the external power line, is an important information that affects the time to find the actual location, thus eliminating the cause of the fault and the overall downtime. At present, we can most accurately determine the location of the fault using a wave locator - impedance principle. The main disadvantage of existing fault locators is that they are based on the assumption of ideally faults takes into account the consideration of symmetrical components of the line. These locators do not take into account the topology of the line, i.e. that the impedances of the individual phases, resp. fault loops are different. This paper proposes, an innovative method of symmetrical components or the Fourier thesis, by using modern informatics and communication tools, which provide opportunities to set new standard in finding the location of the fault on overhead lines in distribution network. Armed with a more precise determination of the failure location, we can reduce the time needed to eliminate the failures and thus, achieve improvements in the reliability of the power distribution system.

Keywords: symmetrical components; fault location

Abstract: Emerging digital technologies enable capturing of a product’s digital footprint through continuous monitoring of its performance, usage, and working environment while using data acquired by its embedded sensors. However, it seems that product development (PD) teams, within engineering companies, have not yet embraced the usage of sensor data acquired by smart products (SPs) when conducting PD activities. This study discusses several challenges that hinder a broader utilization of SP's sensor data, within PD. In addition to the literature review, the discussion is supported with empirical data gathered through a qualitative exploratory case study conducted in a large engineering and manufacturing company. As a result, three challenges are outlined. First, it is challenging for the company's management and PD team to gain transparency over the benefits of sensor data utilization for PD. Second, solutions providing accessibility and visualization of gathered data should be tailored to the PD activities and teams, which requires a holistic understanding. Finally, it is suggested that new skills, roles and processes should be introduced, in order to enable SP's sensor data utilization, within PD activities.

Keywords: smart product; sensor data; product development; digital twin; data-driven design

Abstract: Lean management has become an essential feature for the operation of industrial organizations. Measuring and analyzing lean manufacturing efficiency is a special and complex task. Management control systems should provide feedback on the degree of leanness for effective management and development. In our research we develop a lean manufacturing measurement model based on fuzzy logic, with a management control aspect. We build our model along lean KPIs and value streams. In our research, we highlight through a case study that a fuzzy controlling system can be an effective methodology for measuring leanness. Our results illustrate that the classification of different indicators, using different standardization norms, is not clear. During the analysis of the examined organization, the classification of several lean KPIs and value streams also changed, by changing the standardized norm. With this result, we point out that the definition of intervention points is not clear. The controlling aspect leanness index, developed in our research, allows us to express the expected performance of leanness goals as a function of organizational goals.

Keywords: modelling; controlling; leanness index; lean management; fuzzy logic

Abstract: The flexible job shop problem (FJSP) represents an extension of the classical job shop problem (JSP). The paper deals with a FJSP in an available set of machines with additional transportation time between machines. This type of problem belongs to the group of NP-hard problems. To solve the FJSP, artificial intelligence was used by applying three improved metaheuristic algorithms: Particle Swarm Optimization (PSO) algorithm, Artificial Bee Colonies (ABC) algorithm and Genetic Algorithm (GA). The new approach in solving planning and scheduling problems with additional transportation time in combination with artificial intelligence and the developed neuro-fuzzy system represents the main research subject in the paper. The aim of the paper is to reduce the objective function with regard to time and increase productivity. Based on the case study optimization, experimental results show that the proposed mathematical model and the metaheuristic algorithms lead to an efficient outcome.

Keywords: flexible job shop problem; transportation time; metaheuristic algorithms; anfis; machine learning

Abstract: During the course of personality development, a child’s personality is constantly changing. Adult character traits are a result of these changes. Characteristics such as motivation, conscientiousness, diligence, work capacity and cooperativeness, change and develop throughout the years. A competitive attitude is one of the important qualities in later work-life, the way the employee relates to competitive situations within the organization and what their opinions on these situations are. The present study is the result of research carried out by the authors aiming to examine the correlations between childhood personality traits and later attitudes to workplace competition. The results show that the individual players’ attitude, has an effect on workplace performance and whether the person is result-driven, and that home life, including familial expectations, also influences the intro- and extroversive direction of personality development.

Keywords: Motivation; Competition Attitude; Personality; Workplace Performance; Life Stages