Coordinated inauthentic behavior and information spreading on Twitter

2022 | Decision Support Systems | Citations: 0

Authors: Cinelli, Matteo; Cresci, Stefano; Quattrociocchi, Walter; Tesconi, Maurizio; Zola, Paola

Abstract: We explore the effects of coordinated users (i.e., users characterized by an une ... Expand

Abstract: We explore the effects of coordinated users (i.e., users characterized by an unexpected, suspicious, or exceptional similarity) in information spreading on Twitter by quantifying the efficacy of their tactics in deceiving feed algorithms to maximize information outreach. In particular, we investigate the behavior of coordinated accounts within a large set of retweet-based information cascades identifying key differences between coordinated and non-coordinated accounts in terms of position within the cascade, action delay and outreach. On average, co­ ordinated accounts occupy higher positions of the information cascade (i.e., closer to the root), spread messages faster and involve a slightly higher number of users. When considering cascade metrics such as size, number of edges and height, we observe clear differences among information cascades that are associated to a systemati­ cally larger proportion of coordinated accounts, as confirmed by comparisons with statistical null models. To further characterize the activity of coordinated accounts we introduce two new measures capturing their infectivity within the information cascade (i.e., their ability to involve other users) and their interaction with non-coordinated accounts. Finally, we find that the interaction pattern between the two classes of users follows a saturation-like process. A larger-scale targeting of non-coordinated users does not require a larger amount of coordinated accounts after a threshold value ~50%, after which involving more coordinated accounts within a cascade yields a null marginal effect. Our results contribute to shed light on the role of coordinated accounts and their effect on information diffusion. Collapse

Topics: Twitter social network decision support system application programming interface online learning system

Methods: computational algorithm term frequency–inverse document frequency descriptive statistic linear least square literature study

Mobile Broadband for Inclusive Connectivity: What Deters the High-Capacity Deployment of 4G-LTE Innovation in India?

2022 | Information Systems Frontiers | Citations: 0

Authors: Jha, Ashutosh; Saha, Debashis

Abstract: The global market uptake of the fourth generation-Long-Term Evolution (4G-LTE) M ... Expand

Abstract: The global market uptake of the fourth generation-Long-Term Evolution (4G-LTE) Mobile Broadband (MoBro) services are steadily rising, leading to higher capital investments by mobile network operators (MNOs) to scale-up their infrastructure for meeting the impending demand for MoBro data the world over. However, MNOs face uncertainties about financial returns from such investments, owing to a host of technological and market-related factors, which impact the enablement of such 4G-LTE MoBro services. These challenges are clearly evident in the Indian market. Firstly, 4G-LTE subscribers in the rural India contribute to less than one-third of the overall market size. Secondly, India performs poorly in terms of minimum capacity requirements of the 4G-LTE MoBro services. Thirdly, the rise in per-user consumption of MoBro data does not translate into a similar rise in the MNOs’ revenue. Fourthly, the socio-economic-geographic segregation of India into twenty-two administrative zones (referred to as telecom circles) add to the complexities in the capital investment decisions of MNOs. To address the above challenges, we model various cost and profitability scenarios of a hypothetical MNO providing universal 4G-LTE deployment across the twenty-two telecom circles in India using the available spectrum bands. Our proposed model is firmly established in the “network investment economics” framework for telecom innovations. We adopt a technology diffusion-based approach to forecast the 4G-LTE subscribers in India. We focus on the requirements of 4G-LTE MoBro infrastructure investments, including the spectrum selection decisions by MNOs, and show that, for valuation of the spectrum, the policymakers in India need to take into account the potential of the spectrum in terms of financial returns to an MNO deploying 4G-LTE using that spectrum, as against merely valuing the spectrum based on primarily the technical characteristics of its carrier frequency and benchmarks of prior spectrum auction prices in a particular telecom circle. Finally, we also show that a nationwide 4G-LTE network, which is universal, inclusive, and adhering to the global standards in terms of service quality, can be financially lucrative for MNOs, if enabled by appropriate policies instituting collaborative frameworks for infrastructure sharing, and price rationalization of spectrum bands across the twenty-two telecom circles. Collapse

Topics: mobile system broadband cellular network network infrastructure service quality innovation management

Methods: sensitivity analysis theory development descriptive statistic longitudinal research linear least square

Visual Validation of Estimated Parametric Models of Power Systems Under Ambient Conditions

2019 | HICSS | Citations: 0

Authors: Agrawal, Urmila; Pierre, John W

Abstract: With the advent of synchrophasors, a number of measurement–based algorithms have ... Expand

Abstract: With the advent of synchrophasors, a number of measurement–based algorithms have been developed to estimate system modes using ambient data. The accuracy of these mode estimates depends on a number of factors, such as selection of the system order and data channels. To validate these mode estimates, two methods are proposed in this paper based on how well they agree with the observed data using the estimated parametric system model and observed data. In the proposed methods, the contribution of each mode estimate is analyzed to obtain a combination of system mode estimates to represent the model. The methodology of the proposed methods are illustrated using least–squares ARMA (LS–ARMA) method, which gives a parametric system model estimates. Results obtained by implementing the proposed methods on measured and simulated data validate the effectiveness of the proposed methods. Collapse

Topics: missing data

Methods: autoregressive moving average model simulation computational algorithm linear least square

Wearable Sensor Based Gait Asymmetry Visualization Tool

2018 | Americas Conference on Information Systems | Citations: 0

Authors: Anwary, Arif Reza; Yu, Hongnian; Vassallo, Michael

Abstract: Real time visualization of gait asymmetry can provide added value in rehabilitat ... Expand

Abstract: Real time visualization of gait asymmetry can provide added value in rehabilitation, clinics and sports. Common approaches for the quantification of gait asymmetry give the numerical values of parameters such as symmetry index, symmetry ratio, symmetry angle etc. It may be difficult for users to understand those numerical values. In order to conveniently use quantitative gait asymmetry monitoring for users, an affordable visualization tool is useful to provide a facility for their use in clinic and at home. This paper investigates four approaches for monitoring gait asymmetry to provide automatic graphical visualizations of information about gait. The results show that affordable wearable Inertial Measurement Unit sensors can be used for objective gait asymmetry feature extraction without the requirement and expense of an elaborate laboratory setup. Our procedure significantly simplifies the monitoring protocols and opens possibilities for home based assessment and supports digital transformation strategies through the development of new technology. Collapse

Topics: smart wearable device data visualization mobile application digital transformation

Methods: wearables based research linear least square Shapiro-Wilk test descriptive statistic logarithmic transformation

DIFFUSION AND FORECAST OF MOBILE SERVICE GENERATIONS IN GERMANY, UK, FRANCE AND ITALY – A COMPARATIVE ANALYSIS BASED ON BASS, GOMPERTZ AND SIMPLE LOGISTIC GROWTH MODELS

2018 | European Conference On Information Systems | Citations: 0

Authors: Jha, Ashutosh; Saha, Debashis

Abstract: Growth models, based on the theory of diffusion of innovations, are highly prof ... Expand

Abstract: Growth models, based on the theory of diffusion of innovations, are highly proficient in developing an empirical understanding of country-wide diffusion of mobile services. The currently available literature lacks in explanation of the diffusion of successive generations (G's) of mobile services in various countries. This study furthers the research by analyzing the diffusion of 2G through 4G in Germany, UK, France and Italy, the four largest economies of Europe. We select Bass, Gompertz and Simple Logistic growth models, to analyze the diffusion process, and forecast the adoption of 3G, 4G and 5G mobile broadband, in the four countries. A comparative analysis of the diffusion model parameters, and the forecasting accuracies, estimated through non-linear least-square regression, determines Gompertz and Simple Logistic model as best suited to explain 3G and 4G diffusion, and Bass model as best suited to explain 2G diffusion. Market potential for 3G, 4G and 5G is the highest in France, Germany and Italy, respectively. However, subscribers are more likely to make a direct jump from 2G to 4G in Italy and Germany, compared to UK and France where the gradual switch from 3G to 4G to 5G is likely to be much slower. Collapse

Topics: mobile system logistics management innovation management database system diffusion of innovation

Methods: linear least square cross sectional research theory development mathematical model ordinary least square

Situational Factors Influencing Customers´ Credit Use Online: A Behavioral Economic Approach

2018 | HICSS | Citations: 0

Authors: Øverby Markussen, Ida Merete; Fagerstrøm, Asle; Sydnes, Lars

Abstract: This study investigates consumers´ credit use online from the perspective of in ... Expand

Abstract: This study investigates consumers´ credit use online from the perspective of intertemporal choice and focuses on the impact of personalized credit information when choosing utilitarian versus hedonic product. In a simulated shopping experiment, participants from a Norwegian university college could either save money for the product and get it in the future or buy the product on credit and get it now. A between-group design was used with a randomized selection divided into two groups. The test group (n=37) received personalized credit information while choosing the utilitarian and hedonic products. The control group (n=36) did not have this information. Area Under Curve was calculated and used to make statistical operations. Results show that all participants discounted the saving alternative when the time delay increased, which, therefore, increased their willingness to buy on credit online. Participants´ discounting of the saving alternative was near the hyperbolic model. Second, a significant difference between the utilitarian versus the hedonic products was found for all participants´ willingness to buy on credit online. Finally, personalized information about credit debt had little influence on credit use, but some indications related to hedonic product calls for further research. Implications for research and practice as well as suggestions for future studies are given. Collapse

Methods: experiment machine learning experimental group linear least square statistical hypothesis test

Realizing Value from Business Analytics Platforms: The Effects of Managerial Search and Agility of Resource Allocation Processes

2016 | International Conference on Information Systems | Citations: 9

Authors: Anand, Abhijith; Sharma, Rajeev; Coltman, Tim

Abstract: Few organizations have been able to realize value from their investments in busi ... Expand

Abstract: Few organizations have been able to realize value from their investments in business analytics. This could be due to an inadequate understanding of the pattern of investments required to realize value from business analytics. Specifically, we propose that business analytics requires an upfront investment of infrastructure capital to build a mature platform, followed by multiple investments of innovation capital to create value through competitive actions informed by analytics-enabled insights. Drawing on dynamic capabilities and digital options literatures, we develop a model in which the effect of investments in maturity of the business analytics platform on organizational value is moderated by the agility of the process allocating resources for innovation, and by the efforts expended by line managers to search and select the insights. The model is tested on data collected from a survey of line managers. The findings support the proposed model. Implications and plans for future research are discussed. Collapse

Topics: analytical information system resource allocation organizational value innovation management value creation

Methods: survey self reported survey descriptive statistic statistical hypothesis test literature study

Theories: dynamic capabilities theory measurement theory

Cost Efficiency of Hybrid Cloud Storage: Shortening Acquisition Cycle to Mitigate Volume Variation

2015 | International Conference on Information Systems | Citations: 0

Authors: Laatikainen, Gabriella; Mazhelis, Oleksiy; Tyrväinen, Pasi

Abstract: Hybrid cloud storage infrastructure, which combines cost-effective but inflexibl ... Expand

Abstract: Hybrid cloud storage infrastructure, which combines cost-effective but inflexible private resources and flexible but premium-priced public cloud storage, allows organizations to operate cost-efficiently under demand volume uncertainty. The extant literature, however, offers a limited analytical insight into the effect that the variation of demand has on the cost-efficient mix of internal and external resources. This paper considers the storage capacity acquisition cycle, i.e. the interval at which the organization re-assesses and acquires additional resources, as a parameter shaping the optimal mix of resources. It introduces a model capturing the compound effect of the acquisition cycle and volume variation on the cost-efficiency of hybrid cloud storage. The model is analytically investigated to demonstrate its inherent regularities, and empirically evaluated in numerical experiments. The analysis indicates that shortening the acquisition cycle reduces the volume variability thus reducing the costs. The costs decrease further if shortening the cycle reduces the demand volume uncertainty. Collapse

Topics: cloud storage cloud computing IS sourcing price management organizational transformation

Methods: experiment simulation mathematical model linear least square design science

Theories: transaction cost economics

Preprocessing unbalanced data using support vector machine

2012 | Decision Support Systems | Citations: 9

Authors: Farquad, M. A. H.; Bose, Indranil

Abstract: This paper deals with the application of support vector machine (SVM) to deal wi ... Expand

Abstract: This paper deals with the application of support vector machine (SVM) to deal with the class imbalance problem. The objective of this paper is to examine the feasibility and efficiency of SVM as a preprocessor. Our study analyzes different classification algorithms that are employed to predict the customers with caravan car policy based on his/her sociodemographic data and history of product ownership. A series of experiments was conducted to test various computational intelligence techniques viz., Multilayer Perceptron (MLP), Logistic Regression (LR), and Random Forest (RF). Various standard balancing techniques such as under-sampling, over-sampling and Synthetic Minority Over-sampling TEchnique (SMOTE) are also employed. Subsequently, a strategy of data balancing for handling imbalanced distribution in data is proposed. The proposed approach first employs SVM as a preprocessor and the actual target values of training data are then replaced by the predictions of trained SVM. Later, this modified training data is used to train techniques such as MLP, LR, and RF. Based on the measure of sensitivity, it is observed that the proposed approach not only balances the data effectively but also provides more number of instances for minority class, which in turn enhances the performance of the intelligence techniques. Collapse

Methods: support vector machine computational algorithm experiment machine learning random forest classification

Sales forecasting for computer wholesalers: A comparison of multivariate adaptive regression splines and artificial neural networks

2012 | Decision Support Systems | Citations: 9

Authors: Lu, Chi-Jie; Lee, Tian-Shyug; Lian, Chia-Mei

Abstract: Artificial neural networks (ANNs) have been found to be useful for sales/demand ... Expand

Abstract: Artificial neural networks (ANNs) have been found to be useful for sales/demand forecasting. However, one of the main shortcomings of ANNs is their inability to identify important forecasting variables. This study uses multivariate adaptive regression splines (MARS), a nonlinear and non-parametric regression methodology, to construct sales forecasting models for computer wholesalers. Through the outstanding variable screening ability of MARS, important sales forecasting variables for computer wholesalers can be obtained to enable them to make better sales management decisions. Two sets of real sales data collected from Taiwanese computer wholesalers are used to evaluate the performance of MARS. The experimental results show that the MARS model outperforms backpropagation neural networks, a support vector machine, a cerebellar model articulation controller neural network, an extreme learning machine, an ARIMA model, a multivariate linear regression model, and four two-stage forecasting schemes across various performance criteria. Moreover, the MARS forecasting results provide useful information about the relationships between the forecasting variables selected and sales amounts through the basis functions, important predictor variables, and the MARS prediction function obtained, and hence they have important implications for the implementation of appropriate sales decisions or strategies. Collapse

Topics: sales management

Methods: autoregressive integrated moving average artificial neural network computational algorithm general multivariate regression model Wilcoxon signed-rank test