Skip to main content
Disciplines
Automotive Business IT + Informatics Construction + Real Estate Electrical Engineering + Electronics Energy + Sustainability Insurance + Risk Finance + Banking Management + Leadership Marketing + Sales Mechanical Engineering + Materials
Events
DE
EN
Books
Journals
Topic Page
Marketing
Start single access now
Access for companies
EXTENDED SEARCH
Log in
Top

2019 | Book

Trends in CyberTurfing in the Era of Big Data Read chapter Read first chapter

Business Information Systems

22nd International Conference, BIS 2019, Seville, Spain, June 26–28, 2019, Proceedings, Part II

Editors: Dr. Witold Abramowicz, Rafael Corchuelo

Publisher: Springer International Publishing

Book Series : Lecture Notes in Business Information Processing

Part of: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik" , Springer Professional "Wirtschaft"

Login to get access
insite
SEARCH

About this book

The two-volume set LNBIP 353 and 354 constitutes the proceedings of the 22nd International Conference on Business Information Systems, BIS 2019, held in Seville, Spain, in June 2019.
The theme of the BIS 2019 was "Data Science for Business Information Systems", inspiring researchers to share theoretical and practical knowledge of the different aspects related to Data Science in enterprises. The 67 papers presented in these proceedings were carefully reviewed and selected from 223 submissions.

The contributions were organized in topical sections as follows:

Part I: Big Data and Data Science; Artificial Intelligence; ICT Project Management; and Smart Infrastructure.
Part II: Social Media and Web-based Systems; and Applications, Evaluations and Experiences.

Table of Contents

Frontmatter

Social Media and Web-Based Systems

Frontmatter
Trends in CyberTurfing in the Era of Big Data
Abstract
Previous research on CyberTurfing has been scattered and fragmented in terms of methods and terminology. This paper presents a review of published research related to CyberTurfing in the era of big data. Our objectives were to identify the important terms in this domain and extract essential knowledge from previous studies. We also sought to gain an overview of the trends in CyberTurfing to provide guidance for subsequent research in this field.
Hsiao-Wei Hu, Chia-Ning Wu, Yun Tseng
Keyword-Driven Depressive Tendency Model for Social Media Posts
Abstract
People are increasingly sharing posts on social media (e.g., Facebook, Twitter, Instagram) that include references to their moods/feelings pertaining to their daily lives. In this study, we used sentiment analysis to explore social media messages for hidden indicators of depression. In cooperation with domain experts, we defined a tendency towards depression as evidenced in social media messages based on DSM-5, a standard classification of mental disorders widely used in the U.S. We also developed three data engineering procedures for the extraction of keywords from posts presenting a depressive tendency. Finally, we created a keyword-driven depressive tendency model by which to detect indications of depression in posts on a major social media platform in Taiwan (PTT). The performance of the proposed model was evaluated using three keyword extraction procedures. The DSM-5-based procedure with manual filtering resulted in the highest accuracy (0.74).
Hsiao-Wei Hu, Kai-Shyang Hsu, Connie Lee, Hung-Lin Hu, Cheng-Yen Hsu, Wen-Han Yang, Ling-yun Wang, Ting-An Chen
Exploring Interactions in Social Networks for Influence Discovery
Abstract
Today’s social networks allow users to react to new contents such as images, posts and messages in numerous ways. For example, a user, impressed by another user’s post, might react to it by liking it and then sharing it forward to her friends. Therefore, a successful estimation of the influence between users requires models to be expressive enough to fully describe various reactions. In this article, we aim to utilize those direct reactive activities, in order to calculate users impact on others. Hence, we propose a flexible method that considers type, quality, quantity and time of reactions and, as a result, the method assesses the influence dependencies within the social network. The experiments conducted using two different real-world datasets of Facebook and Pinterest show the adequacy and flexibility of the proposed model that is adaptive to data having different features.
Monika Ewa Rakoczy, Amel Bouzeghoub, Katarzyna Wegrzyn-Wolska, Alda Lopes Gancarski
A Literature Review on Application Areas of Social Media Analytics
Abstract
The use of social media is part of everyday life in both private and professional environments. Social media is used for communication, data exchange and the distribution of news and advertisements. Social Media Analytics (SMA) help to collect and interpret unstructured data. The measurement of user behavior serves to form opinions and evaluate the influence of individual actors. This results in a multitude of application areas for SMA. On the basis of a literature search, our aim is to determine the main application areas and summarize the current state of research. We describe these areas, show current findings from the literature and uncover gaps in research. The main application areas of SMA investigated in research are healthcare, tourism and natural disaster control.
Kirsten Liere-Netheler, León Gilhaus, Kristin Vogelsang, Uwe Hoppe
Personalized Cloud Service Review Ranking Approach Based on Probabilistic Ontology
Abstract
Online cloud service reviews have recently gained an increasing attention since they can have a significant impact on cloud user’ purchasing decision. A large number of cloud users consult these reviews before choosing cloud services. Therefore, identifying the most-helpful reviews is an important task for online retailers. The helpfulness of product/service reviews has been widely investigated in the marketing domain. However, these works do not pay attention to the following significant points: (1) the heterogeneity problem when extracting information from different Social Media Platforms (SMP), (2) the uncertainty judgment of review helpfulness and (3) the personalizing of review ranking by considering the context of the review. To tackle these three points we propose a new approach that relies on probabilistic ontology, called Context-aware Review Helpfulness Probabilistic Ontology (C-RHPO), to cope with the heterogeneity and uncertainty issues. In addition, the approach uses a personalized online review ranking method based on the end-user context. The herein reported experimental results proved the effectiveness and the performance of the approach.
Emna Ben-Abdallah, Khouloud Boukadi, Mohamed Hammami
Influential Nodes Detection in Dynamic Social Networks
Abstract
The influence maximization problem aims to identify influential nodes allowing to reach the viral marketing objectives on social networks. Previous researches are mainly concerned with the static social network analysis and the development of algorithms in this context. However, when network changes, those algorithms must be updated. In this paper, we offer a new interesting approach to study the influential nodes detection problem in changing social networks. This approach can be considered to be an extension of a previous static algorithm SND (Semantic and structural influential Nodes Detection). Experimental results prove the effectiveness of SNDUpdate to detect influential nodes in dynamic social networks.
Nesrine Hafiene, Wafa Karoui, Lotfi Ben Romdhane
A Fuzzy Modeling Approach for Group Decision Making in Social Networks
Abstract
Social networks have been commonly used, people use social networks with various purposes, such as, enjoying time, making business, and contacting their friends. All these activities are mainly based on sharing data. In social networks, making decision on data sharing process has become one of the main challenge because it involves people who have different opinions on the same problem. Diversified opinions cause uncertainties in decision making process. Fuzzy logic is used to overcome uncertainties’ situations. In this work, we provide a fuzzy logic based decision making framework for SNs. The proposed fuzzy logic based framework uses data sensitivity value and trust value (confidence value) to make the group decision. Users express their opinions on data security features to obtain aggregated decision. Facebook data sharing process is chosen as a case study.
Gulsum Akkuzu, Benjamin Aziz, Mo Adda
System Modeling by Representing Information Systems as Hypergraphs
Abstract
Hypergraph as a formal model offers a sound foundation for representing information systems. There are several issues that are worth observing during analysis, design, and operation of information systems such as consistency, integrity, soundness of control and security mechanisms. The improvement and advancement of machine learning and data science algorithms provide the opportunity to spot patterns, to predict and to prescript some activities within complex environments that can depict huge sets of data. Our proposal is that the available algorithms can be applied on hypergraphs through profound customization whereby the capability of algorithms can be exploited for Business Information Systems.
Bence Sarkadi-Nagy, Bálint Molnár
Development of a Social Media Maturity Model for Logistics Service Providers
Abstract
Logistics service providers (LSPs) conduct their business in an environment of steadily changing stakeholders and business models. Social media (SM) has become an important communication tool and source for new business models for LSPs. Nevertheless, a lot of LSPs struggle with the utilization of SM. In this paper, we develop an SM maturity model (MM) for LSPs. By doing so, our research sheds light on the use of SM at LSPs and reveals impediments. Thus, the developed MM will help researchers better understand the utilization of SM at LSPs and practitioners to improve their business processes.
Axel Jacob, Frank Teuteberg
Potential Benefits of New Online Marketing Approaches
Abstract
This study examines the potential benefits of new approaches in online or digital marketing. In the course of this study, the research design and the new approaches in online marketing are considered. In a specially prepared quantitative study, experts were questioned about the individual approaches by means of a questionnaire. The questionnaire is based on derived hypotheses from the literature. The study focuses on the analysis of the survey results using the SmartPLS software. After data analysis using structural equation modeling, the results show that Mobile and Data-driven Marketing as well as Programmatic Advertising do have a significant influence on the potential of the new approaches in online marketing. The results are used to recommend actions for enterprises.
Ralf-Christian Härting, Christopher Reichstein, Andreas Müller

Applications, Evaluations and Experiences

Frontmatter
Using Blockchain Technology for Cross-Organizational Process Mining – Concept and Case Study
Abstract
Business processes in companies lead to an enormous number of event logs in their IT systems. Evaluating these event logs using data mining can provide companies with valuable process analysis information which can uncover process improvement potentials. However, media breaks frequently occur in these processes, so that there is a risk of optimizing isolated sub-processes only. Blockchain technology may avoid these media breaks and thus create the basis for complete event log analysis. The focus of our paper is to investigate existing requirements and to identify a blockchain based solution scenario evaluated by experts.
Stefan Tönnissen, Frank Teuteberg
Modeling the Cashflow Management of Bike Sharing Industry
Abstract
The sharing economy has been widely concerned since its inception and bike sharing is one of the most representative examples. The paper attempts to investigate the cashflow management strategy of bike sharing companies to optimize the overall financial return. The framework of our model is based on the assumption that bike sharing companies may invest operation income in financial market for long-term and short-term earnings. Optimal reserve pool is modeled and estimated using double parameterized compound Poisson distributions. Empirical examples and Monte Carlo analysis are provided for model validation.
Binrui Shen, Yu Shan, Yunyu Jia, Dejun Xie, Shengxin Zhu
Predicting Material Requirements in the Automotive Industry Using Data Mining
Abstract
Advanced capabilities in artificial intelligence pave the way for improving the prediction of material requirements in automotive industry applications. Due to uncertainty of demand, it is essential to understand how historical data on customer orders can effectively be used to predict the quantities of parts with long lead times. For determining the accuracy of these predications, we propose a novel data mining technique. Our experimental evaluation uses a unique, real-world data set. Throughout the experiments, the proposed technique achieves high accuracy of up to 98%. Our research contributes to the emerging field of data-driven decision support in the automotive industry.
Tobias Widmer, Achim Klein, Philipp Wachter, Sebastian Meyl
Are Similar Cases Treated Similarly? A Comparison Between Process Workers
Abstract
In processes involving human professional judgment (e.g., in Knowledge Intensive processes) it is not easy to verify if similar cases receive similar treatment. In these processes there is a risk of dissimilar treatment as human process workers may develop their individual experiences and convictions or change their behavior due to changes in workload or season. Awareness of dissimilar treatment of similar cases may prevent disputes, inefficiencies, or non-compliance with regulations that require similar treatment of similar cases. In this article two procedures are presented for testing in an objective (statistical) way if different groups of process workers treat similar cases in a similar way. The testing is based on splitting the event log of a process in parts corresponding to the different (groups of) process workers and analyzing the sequences of events in each part. The two procedures are demonstrated on an example using synthetic data and on a real life event log.
Mark Pijnenburg, Wojtek Kowalczyk
Mining Labor Market Requirements Using Distributional Semantic Models and Deep Learning
Abstract
This article describes a new method for analyzing labor market requirements by matching job listings from online recruitment platforms with professional standards to weigh the importance of particular professional functions and requirements and enrich the general concepts of professional standards using real labor market requirements. Our approach aims to combat the gap between professional standards and reality of fast changing requirements in developing branches of economy. First, we determine professions for each job description, using the multi-label classifier based on convolutional neural networks. Secondly, we solve the task of concept matching between job descriptions and standards for the respective professions by applying distributional semantic models. In this task, the average word2vec model achieved the best performance among other vector space models. Finally, we experiment with expanding general vocabulary of professional standards with the most frequent unigrams and bigrams occurring in matching job descriptions. Performance evaluation is carried out on a representative corpus of job listings and professional standards in the field of IT.
Dmitriy Botov, Julius Klenin, Andrey Melnikov, Yuri Dmitrin, Ivan Nikolaev, Mikhail Vinel
Enhancing Supply Chain Risk Management by Applying Machine Learning to Identify Risks
Abstract
Supply chain risks negatively affect the success of an OEM in automotive industry. Finding relevant information for supply chain risk management (SCRM) is a critical task. This investigation utilizes machine learning to find risk within textual documents. It contributes to the supply chain management (SCM) by designing (i) a conceptual model for supply risk identification in textual data. This addresses the requirement to see the direct connection between data analytics and SCM. (ii) An experiment in which a prototype is evaluated contributes the requirement to have more empirical insight in the interdisciplinary field of data analytics in SCRM.
Ahmad Pajam Hassan
Deep Neural Networks for Driver Identification Using Accelerometer Signals from Smartphones
Abstract
With the evolution of the onboard communications services and the applications of ride-sharing, there is a growing need to identify the driver. This identification, within a given driver set, helps in tasks of antitheft, autonomous driving, fleet management systems or automobile insurance. The object of this paper is to identify a driver in the least invasive way possible, using the smartphone that the driver carries inside the vehicle in a free position, and using the minimum number of sensors, only with the tri-axial accelerometer signals from the smartphone. For this purpose, different Deep Neural Networks have been tested, such as the ResNet-50 model and Recurrent Neural Networks. For the training, temporal signals of the accelerometers have been transformed as images. The accuracies obtained have been 69.92% and 90.31% at top-1 and top-5 driver level respectively, for a group of 25 drivers. These results outperform works in the state of the art, which can even utilize more signals (like GPS- Global Positioning System- measurement data) or extra-equipment (like the Controller Area-Network of the vehicle).
Sara Hernández Sánchez, Rubén Fernández Pozo, Luis Alfonso Hernández Gómez
Dynamic Enterprise Architecture Capabilities: Conceptualization and Validation
Abstract
The notion of enterprise architecture (EA) and EA-based capabilities in IS literature has emerged as an important research domain. However, the conceptualizations of EA-based capabilities remain ambiguous, largely not validated and still lack a firm base in theory. This study, therefore, aims to rigorously conceptualize EA-based capabilities grounded in theory and puts forward the notion of dynamic enterprise architecture capabilities. These capabilities highlight the core areas in which organizations should infuse EA. The purpose of this study is to develop a reliable and valid measurement scale. This scale is validated using item-sorting analyses, expert reviews and an empirical study of 299 CIOs and enterprise architects. The outcomes support the validity and reliability of the scale. The dynamic enterprise architecture capabilities scale developed in this research contributes to theory development and the EA knowledge base. The scale may be used as an assessment or benchmarking tool in practice.
Rogier van de Wetering
Re-engineering Higher Education Learning and Teaching Business Processes for Big Data Analytics
Abstract
Big Data analytics need to be combined with higher education business processes to improve course structure and delivery to help students who have struggled to stay in the course by identifying their engagement and correlation with different variables such as access to documents; assignment submission etc. Online activity data can be used to keep students on track all the way to graduation and universities struggling to understand how to lower dropout rates and keep students on track during their study program. In this paper we discuss how Big Data analytics can be combined with higher education business processes using re-engineering for structured data, unstructured data, and external data. In order to achieve this objective, we investigate the core business processes of learning and teaching and define a re-engineered higher education business process model.
Meena Jha, Sanjay Jha, Liam O’Brien
Real-Time Age Detection Using a Convolutional Neural Network
Abstract
The problem of determining people’s age is a recurring theme in areas such as law enforcement, education and sports because age is often used to determine eligibility. The aim of current work is to make use of a lightweight machine learning model for automating the task of detecting people’s age. This paper presents a solution that makes use of a lightweight Convolutional Neural Network model, built according to a modification of the LeNet-5 architecture to perform age detection, for both males and females, in real-time. The UTK-Face Large Scale Face Dataset was used to train and test the performance of the model in terms of predicting age. To evaluate the model’s performance in real-time, Haar Cascades were used to detect faces from video feeds. The detected faces were fed to the model for it to make age predictions. Experimental results showed that age-detection can be performed in real-time. Although, the prediction accuracy of the model requires improvement.
Siphesihle Sithungu, Dustin Van der Haar
An Inventory-Based Mobile Application for Warehouse Management to Digitize Very Small Enterprises
Abstract
In today’s world, digitization has reached an important role. While more and more enterprises are interested in a realization and aware of its importance, it rather lacks on the implementation. This applies most of all for those with a low budget for new investments, such as small and medium enterprises (SME), as well as even smaller forms (VSE). For this reason, a sectoral digitization approach based on a real-world VSE is pursued. By using the design science research methodology, a low cost and easy to use warehouse management system for the optimization of the inner logistics of VSEs is developed, evaluated and presented in this work.
Daniel Staegemann, Matthias Volk, Christian Lucht, Christian Klie, Michael Hintze, Klaus Turowski
Collaboration in Mixed Homecare – A Study of Care Actors’ Acceptance Towards Supportive Groupware
Abstract
As more and more people reach high age the need for care, especially at home, rises. Caring involves the coordination of a wide variety of actors. Modern information and communication technologies (ICT) may improve care coordination and thus relieve all actors involved in outpatient care.
This paper presents the results of a study (n = 108), that aimed to find out about the attitude of care actors towards digital care coordination tools in Germany. The survey contained questions regarding the care situation, expectations, technology commitment, barriers and need for assistance.
The data were primarily evaluated according to the subgroups informal caregivers and professional actors. The study showed a lack of target group oriented provision and support of groupware. A mere provision of the technology does not lead to the desired acceptance of the offer because none of the actor groups sees the initiating role of technology use on their side. Personal instruction and support are in demand in both user groups, regardless of technology commitment. For the rather less technology-savvy informal caregivers, this can be explained through their rather tense care situations and the mostly rather high age and the associated restrictions. Professionals demand to learn the technology in order to integrate it as effectively as possible into their daily care routine.
Madeleine Renyi, Melanie Rosner, Frank Teuteberg, Christophe Kunze
Stress-Sensitive IT-Systems at Work: Insights from an Empirical Investigation
Abstract
High workload, complex and knowledge-intense tasks as well as increased expectations in respect to flexibility and timeliness give rise to work intensification. This can lead to permanent stress causing serious health problems. Thus, it is a major concern to take measures against stress in order to maintain workers’ health and productivity. While information technology provides great potential to mitigate work-induced stress, preferences of workers regarding IT-based assistance are largely unknown. Against this research gap, we conducted a quantitative study on the acceptance and feasibility of implementation options for stress-sensitive systems. Our results are intended to inform future research in the design and development of such systems.
Michael Fellmann, Fabienne Lambusch, Anne Waller
Data Quality Management Framework for Smart Grid Systems
Abstract
New devices in smart grid such as smart meters and sensors have emerged to become a massive and complex network, where a large volume of data is flowing to the smart grid systems. Those data can be real-time, fast-moving, and originated from a vast variety of terminal devices. However, the big smart grid data also bring various data quality problems, which may cause the delayed, inaccurate analysis of results, even fatal errors in the smart grid system. This paper, therefore, identifies a comprehensive taxonomy of typical data quality problems in the smart grid. Based on the adaptation of established data quality research and frameworks, this paper proposes a new data quality management framework that classifies the typical data quality problems into related data quality dimensions, contexts, as well as countermeasures. Based on this framework, this paper not only provides a systematic overview of data quality in the smart grid domain, but also offers practical guidance to improve data quality in smart grids such as which data quality dimensions are critical and which data quality problems can be addressed in which context.
Mouzhi Ge, Stanislav Chren, Bruno Rossi, Tomas Pitner
Balancing Performance Measures in Classification Using Ensemble Learning Methods
Abstract
Ensemble learning methods have recently been widely used in various domains and applications owing to the improvements in computational efficiency and distributed computing advances. However, with the advent of wide variety of applications of machine learning techniques to class imbalance problems, further focus is needed to evaluate, improve and balance other performance measures such as sensitivity (true positive rate) and specificity (true negative rate) in classification. This paper demonstrates an approach to evaluate and balance the performance measures (specifically sensitivity and specificity) using ensemble learning methods for classification that can be especially useful in class imbalanced datasets. In this paper, ensemble learning methods (specifically bagging and boosting) are used to balance the performance measures (sensitivity and specificity) on a diabetes dataset to predict if a patient will be readmitted to the hospital based on various feature vectors. From the experiments conducted, it can be empirically concluded that, by using ensemble learning methods, although accuracy does improve to some margin, both sensitivity and specificity are balanced significantly and consistently over different cross validation approaches.
Neeraj Bahl, Ajay Bansal
Machine Learning for Engineering Processes
Abstract
Buildings are realized through engineering processes in projects, that however tend to result in cost and/or time overrun. Therefore, a need is highlighted by the industry and the literature, to develop predictive models, that can aid in decision-making and guidance, especially in a preparation effort before production is initiated.
This study aims at investigating what are possible applications of machine learning in building engineering projects and how they impact on their performance?
First, a literature review about machine learning (ML) is done. The first case is drawing on a productivity survey of building projects in Sweden (n = 580). The most influential factors behind project performance are identified, to predict performance. Features that are strongly correlated with four performance indicators are identified: cost variance, time variance and client- and contractor satisfaction and a regression analysis is done. Human related factors predict success best, such as the client role, the architect performance and collaboration. But external factors and technical aspects of a building are also important.
The second case combines constructability and risk analysis on a basis on civil engineering project from several different countries and with very different character; a town square, a biogas plant, road bridges and sub projects from an airport. The data encompasses 30 projects. The development build on literature study, expert interview, unsupervised and supervised learning. The strength lies more in the conceptual work of risk sources enabled by ML. Human reasoning is needed in building projects. Also after the introduction of ML.
Christian Koch
Backmatter
Metadata
Title
Business Information Systems
Editors
Dr. Witold Abramowicz
Rafael Corchuelo
Copyright Year
2019
Electronic ISBN
978-3-030-20482-2
Print ISBN
978-3-030-20481-5
DOI
https://doi.org/10.1007/978-3-030-20482-2

Premium Partner

    Image Credits