Search results
1 – 10 of over 1000Charitha Sasika Hettiarachchi, Nanfei Sun, Trang Minh Quynh Le and Naveed Saleem
The COVID-19 pandemic has posed many challenges in almost all sectors around the globe. Because of the pandemic, government entities responsible for managing health-care resources…
Abstract
Purpose
The COVID-19 pandemic has posed many challenges in almost all sectors around the globe. Because of the pandemic, government entities responsible for managing health-care resources face challenges in managing and distributing their limited and valuable health resources. In addition, severe outbreaks may occur in a small or large geographical area. Therefore, county-level preparation is crucial for officials and organizations who manage such disease outbreaks. However, most COVID-19-related research projects have focused on either state- or country-level. Only a few studies have considered county-level preparations, such as identifying high-risk counties of a particular state to fight against the COVID-19 pandemic. Therefore, the purpose of this research is to prioritize counties in a state based on their COVID-19-related risks to manage the COVID outbreak effectively.
Design/methodology/approach
In this research, the authors use a systematic hybrid approach that uses a clustering technique to group counties that share similar COVID conditions and use a multi-criteria decision-making approach – the analytic hierarchy process – to rank clusters with respect to the severity of the pandemic. The clustering was performed using two methods, k-means and fuzzy c-means, but only one of them was used at a time during the experiment.
Findings
The results of this study indicate that the proposed approach can effectively identify and rank the most vulnerable counties in a particular state. Hence, state health resources managing entities can identify counties in desperate need of more attention before they allocate their resources and better prepare those counties before another surge.
Originality/value
To the best of the authors’ knowledge, this study is the first to use both an unsupervised learning approach and the analytic hierarchy process to identify and rank state counties in accordance with the severity of COVID-19.
Details
Keywords
Sunil Kumar Jauhar, B. Ripon Chakma, Sachin S. Kamble and Amine Belhadi
As e-commerce has expanded rapidly, online shopping platforms have become widespread in India and throughout the world. Product return, which has a negative effect on the…
Abstract
Purpose
As e-commerce has expanded rapidly, online shopping platforms have become widespread in India and throughout the world. Product return, which has a negative effect on the E-Commerce Industry's economic and ecological sustainability, is one of the E-Commerce Industry's greatest challenges in light of the substantial increase in online transactions. The authors have analyzed the purchasing patterns of the customers to better comprehend their product purchase and return patterns.
Design/methodology/approach
The authors utilized digital transformation techniques-based recency, frequency and monetary models to better understand and segment potential customers in order to address personalized strategies to increase sales, and the authors performed seller clustering using k-means and hierarchical clustering to determine why some sellers have the most sales and what products they offer that entice customers to purchase.
Findings
The authors discovered, through the application of digital transformation models to customer segmentation, that over 61.15% of consumers are likely to purchase, loyal customers and utilize firm service, whereas approximately 35% of customers have either stopped purchasing or have relatively low spending. To retain these consumer segments, special consideration and an enticing offer are required. As the authors dug deeper into the seller clustering, we discovered that the maximum number of clusters is six, while certain clusters indicate that prompt delivery of the goods plays a crucial role in customer feedback and high sales volume.
Originality/value
This is one of the rare study that develops a seller segmentation strategy by utilizing digital transformation-based methods in order to achieve seller group division.
Details
Keywords
Nihan Yildirim, Derya Gultekin, Cansu Hürses and Abdullah Mert Akman
This paper aims to use text mining methods to explore the similarities and differences between countries’ national digital transformation (DT) and Industry 4.0 (I4.0) policies…
Abstract
Purpose
This paper aims to use text mining methods to explore the similarities and differences between countries’ national digital transformation (DT) and Industry 4.0 (I4.0) policies. The study examines the applicability of text mining as an alternative for comprehensive clustering of national I4.0 and DT strategies, encouraging policy researchers toward data science that can offer rapid policy analysis and benchmarking.
Design/methodology/approach
With an exploratory research approach, topic modeling, principal component analysis and unsupervised machine learning algorithms (k-means and hierarchical clustering) are used for clustering national I4.0 and DT strategies. This paper uses a corpus of policy documents and related scientific publications from several countries and integrate their science and technology performance. The paper also presents the positioning of Türkiye’s I4.0 and DT national policy as a case from a developing country context.
Findings
Text mining provides meaningful clustering results on similarities and differences between countries regarding their national I4.0 and DT policies, aligned with their geographic, economic and political circumstances. Findings also shed light on the DT strategic landscape and the key themes spanning various policy dimensions. Drawing from the Turkish case, political options are discussed in the context of developing (follower) countries’ I4.0 and DT.
Practical implications
The paper reveals meaningful clustering results on similarities and differences between countries regarding their national I4.0 and DT policies, reflecting political proximities aligned with their geographic, economic and political circumstances. This can help policymakers to comparatively understand national DT and I4.0 policies and use this knowledge to reflect collaborative and competitive measures to their policies.
Originality/value
This paper provides a unique combined methodology for text mining-based policy analysis in the DT context, which has not been adopted. In an era where computational social science and machine learning have gained importance and adaptability to political and social science fields, and in the technology and innovation management discipline, clustering applications showed similar and different policy patterns in a timely and unbiased manner.
Details
Keywords
Josemila Baby Jesuretnam and Jeba James Rose
This paper aims to propose a multi-dimensional hierarchical K-means clustering algorithm for the purpose of intrusion detection. Initially, the clustering set of rules is proposed…
Abstract
Purpose
This paper aims to propose a multi-dimensional hierarchical K-means clustering algorithm for the purpose of intrusion detection. Initially, the clustering set of rules is proposed to shape some of clusters in the network and then the most beneficial clusters are decided on by the use of Cuckoo search optimization set of rules. Finally, an Artificial Bee Colony primarily based selection tree (ABC-DT) classifier is rented to classify the regular and unusual instances present in the network with the aid of the extracted features.
Design/methodology/approach
Intrusion detection system (IDS) is crucial for the network system; the intruder can take sensitive details about the network. IDS are said to be more effective when it has both high intrusion detection rate and low false alarm rate. Numerous strategies including gadget mastering, records mining and statistical techniques were tested for IDS mission. Recent study reveals that combining multiple classifiers, i.e. classifiers ensemble, can also own better performance than unmarried classifier. In this paper, a comparative study is conducted of the overall performance of four classifiers, i.e. hybrid ABC-DT particle swarm optimization-based K-means clustering (PSO-KM), help vector device (SVM) and K-Nearest neighbour (KNN). All the four classifiers are tested with exceptional packet sizes 1470, 1024, 512 and 256. The experiment is carried out for the speed ranging from turned into done for the velocity ranging from 250Mbps, 500Mbps, 750Mbps, 1.0Gpbs, 1.5Gbps, and 2.0Gbps in terms of accuracy, detection charge, specificity, false alarm charge and computational time. The experimental results reveals that the hybridization of classifiers performs better than the base classifiers in all scenarios.
Findings
This study analyses the performance of hybrid ABC-DT classifier and compares the performance against three well-known classifiers such as PSO-KM, SVM and K-NN. The performances of all the four classifiers are tested with Discovery in Data Mining (KDD) CUP 99 dataset with different packet sizes 1470, 1024, 512 and 256. The results show the classifier performance variations with different speed ranges. From the experimental results and analysis, the hybridization of classifiers such as ABC-DT outperforms the base classifiers in all scenarios.
Originality/value
The novel approach in this paper is used to study the hybrid ABC-DT classifier and compare the performance against three well-known classifiers such as PSO-KM, SVM and K-NN. The discussed concept is used within the network to monitor the traffic to and from all the devices connected in that network.
Details
Keywords
Given the recent explosion of interest in streaming data and online algorithms, clustering of time series subsequences has received much attention. In this work we make a…
Abstract
Given the recent explosion of interest in streaming data and online algorithms, clustering of time series subsequences has received much attention. In this work we make a surprising claim. Clustering of time series subsequences is completely meaningless. More concretely, clusters extracted from these time series are forced to obey a certain constraint that is pathologically unlikely to be satisfied by any dataset, and because of this, the clusters extracted by any clustering algorithm are essentially random. While this constraint can be intuitively demonstrated with a simple illustration and is simple to prove, it has never appeared in the literature. We can justify calling our claim surprising, since it invalidates the contribution of dozens of previously published papers. We will justify our claim with a theorem, illustrative examples, and a comprehensive set of experiments on reimplementations of previous work.
Simon Wiersma, Tobias Just and Michael Heinrich
Germany has a polycentric city structure. This paper aims to reduce complexity of this structure and to find a reliable classification scheme of German housing markets at city…
Abstract
Purpose
Germany has a polycentric city structure. This paper aims to reduce complexity of this structure and to find a reliable classification scheme of German housing markets at city level based on 17 relevant market parameters.
Design/methodology/approach
This paper uses a two-step clustering algorithm combining k-means with Ward’s method to develop the classification scheme. The clustering process is preceded by a principal component analysis to merely retain the most important dimensions of the market parameters. The robustness of the results is investigated with a bootstrapping method.
Findings
It is found that German residential markets can best be segmented into four groups. Geographic contiguity plays a specific role, but is not a main factor. Our bootstrapping analysis identifies the majority of pairwise city relations (88.5%) to be non-random.
Research limitations/implications
A deeper discussion concerning the most relevant market parameters is required. The stability of the clusters is to be re-investigated in the future, as the bootstrapping analysis indicates that some clusters are more homogeneous than others.
Practical implications
The developed classification scheme provides insights into opportunities and risks associated with specific city groups. The findings of this study can be used in portfolio management to reduce unsystematic investment risks and to formulate investment strategies.
Originality/value
To the best of the authors’ knowledge, this is the first paper to offer insights into the German housing markets which applies principal component, cluster and bootstrapping analyses in a sole integrated approach.
Details
Keywords
Dharyll Prince Mariscal Abellana and Paula Esplanada Mayol
This paper aims to propose a novel hybrid-decision-making trial and evaluation laboratory-K means clustering algorithm as a decision-making framework for analyzing the barriers of…
Abstract
Purpose
This paper aims to propose a novel hybrid-decision-making trial and evaluation laboratory-K means clustering algorithm as a decision-making framework for analyzing the barriers of green computing adoption.
Design/methodology/approach
A literature review is conducted to extract relevant green computing barriers. An expert elicitation process is performed to finalize the barriers and to establish their corresponding interrelationships.
Findings
The proposed approach offers a comprehensive framework for modeling the barriers of green computing adoption.
Research limitations/implications
The results of this paper provide insights on how the barriers of green computing adoption facilitate the adoption of stakeholders. Moreover, the paper provides a framework for analyzing the structural relationships that exist between factors in a tractable manner.
Originality/value
The paper is one of the very first attempts to analyze the barriers of green computing adoption. Furthermore, it is the first to offer lenses in a Philippine perspective. The paper offers a novel algorithm that can be useful in modeling the barriers of innovation, particularly, in green computing adoption.
Details
Keywords
Hannan Amoozad Mahdiraji, Madjid Tavana, Pouya Mahdiani and Ali Asghar Abbasi Kamardi
Customer differences and similarities play a crucial role in service operations, and service industries need to develop various strategies for different customer types. This study…
Abstract
Purpose
Customer differences and similarities play a crucial role in service operations, and service industries need to develop various strategies for different customer types. This study aims to understand the behavioral pattern of customers in the banking industry by proposing a hybrid data mining approach with rule extraction and service operation benchmarking.
Design/methodology/approach
The authors analyze customer data to identify the best customers using a modified recency, frequency and monetary (RFM) model and K-means clustering. The number of clusters is determined with a two-step K-means quality analysis based on the Silhouette, Davies–Bouldin and Calinski–Harabasz indices and the evaluation based on distance from average solution (EDAS). The best–worst method (BWM) and the total area based on orthogonal vectors (TAOV) are used next to sort the clusters. Finally, the associative rules and the Apriori algorithm are used to derive the customers' behavior patterns.
Findings
As a result of implementing the proposed approach in the financial service industry, customers were segmented and ranked into six clusters by analyzing 20,000 records. Furthermore, frequent customer financial behavior patterns were recognized based on demographic characteristics and financial transactions of customers. Thus, customer types were classified as highly loyal, loyal, high-interacting, low-interacting and missing customers. Eventually, appropriate strategies for interacting with each customer type were proposed.
Originality/value
The authors propose a novel hybrid multi-attribute data mining approach for rule extraction and the service operations benchmarking approach by combining data mining tools with a multilayer decision-making approach. The proposed hybrid approach has been implemented in a large-scale problem in the financial services industry.
Details
Keywords
Khatab Alqararah and Ibrahim Alnafrah
This research paper aims to contribute to the field of innovation performance benchmarking by identifying appropriate benchmarking groups and exploring learning opportunities and…
Abstract
Purpose
This research paper aims to contribute to the field of innovation performance benchmarking by identifying appropriate benchmarking groups and exploring learning opportunities and integration directions.
Design/methodology/approach
The study employs a multi-dimensional innovation-driven clustering methodology to analyze data from the 2019 edition of the Global Innovation Index (GII). Hierarchical and K-means Cluster Analysis techniques are applied using various sets of distance matrices to uncover and analyze distinct innovation patterns.
Findings
This study classifies 129 countries into four clusters: Specials, Advanced, Intermediates and Primitives. Each cluster exhibits strengths and weaknesses in terms of innovation performance. Specials excel in the areas of institutions and knowledge commercialization, while the Advanced cluster demonstrates strengths in education and ICT-related services but shows weakness in patent commercialization. Intermediates show strengths in venture-capital and labour productivity but display weaknesses in R&D expenditure and the higher education quality. Primitives exhibit strength in creative activities but suffer from weaknesses in digital skills, education and training. Additionally, the study has identified 35 indicators that have negligible variance contributions across countries.
Originality/value
The study contributes to finding the relevant countries’ grouping for the enhancement of communication, integration and learning. To this end, this study highlights the innovation structural differences among countries and provides tailored innovation policies.
Details
Keywords
Peiman Alipour Sarvari, Alp Ustundag and Hidayet Takci
The purpose of this paper is to determine the best approach to customer segmentation and to extrapolate associated rules for this based on recency, frequency and monetary (RFM…
Abstract
Purpose
The purpose of this paper is to determine the best approach to customer segmentation and to extrapolate associated rules for this based on recency, frequency and monetary (RFM) considerations as well as demographic factors. In this study, the impacts of RFM and demographic attributes have been challenged in order to enrich factors that lend comprehension to customer segmentation. Different types of scenario were designed, performed and evaluated meticulously under uniform test conditions. The data for this study were extracted from the database of a global pizza restaurant chain in Turkey. This paper summarizes the findings of the study and also provides evidence of its empirical implications to improve the performance of customer segmentation as well as achieving extracted rule perfection via effective model factors and variations. Accordingly, marketing and service processes will work more effectively and efficiently for customers and society. The implication of this study is that it explains a clear concept for interaction between producers and consumers.
Design/methodology/approach
Customer relationship management, which aims to manage record and evaluate customer interactions, is generally regarded as a vital tool for companies that wish to be successful in the rapidly changing global market. The prediction of customer behaviors is a strategically important and difficult issue because of the high variance and wide range of customer orders and preferences. So to have an effective tool for extracting rules based on customer purchasing behavior, considering tangible and intangible criteria is highly important. To overcome the challenges imposed by the multifaceted nature of this problem, the authors utilized artificial intelligence methods, including k-means clustering, Apriori association rule mining (ARM) and neural networks. The main idea was that customer clusters are better enhanced when segmentation processes are based on RFM analysis accompanied by demographic data. Weighted RFM (WRFM) and unweighted RFM values/scores were applied with and without demographic factors and utilized to compose different types and numbers of clusters. The Apriori algorithm was used to extract rules of association. The performance analyses of scenarios have been conducted based on these extracted rules. The number of rules, elapsed time and prediction accuracy were used to evaluate the different scenarios. The results of evaluations were compared with the outputs of another available technique.
Findings
The results showed that having an appropriate segmentation approach is vital if there are to be strong association rules. Also, it has been determined from the results that the weights of RFM attributes affect rule association performance positively. Moreover, to capture more accurate customer segments, a combination of RFM and demographic attributes is recommended for clustering. The results’ analyses indicate the undeniable importance of demographic data merged with WRFM. Above all, this challenge introduced the best possible sequence of factors for an analysis of clustering and ARM based on RFM and demographic data.
Originality/value
The work compared k-means and Kohonen clustering methods in its segmentation phase to prove the superiority of adopted segmentation techniques. In addition, this study indicated that customer segments containing WRFM scores and demographic data in the same clusters brought about stronger and more accurate association rules for the understanding of customer behavior. These so-called achievements were compared with the results of classical approaches in order to support the credibility of the proposed methodology. Based on previous works, classical methods for customer segmentation have overlooked any combination of demographic data with WRFM during clustering before proceeding to their rule extraction stages.
Details