Breakthroughs on Cross-Cutting Data Management, Data Analytics, and Applied Data Science

2021 | Information Systems Frontiers | Citations: 0

Authors: Chiusano, Silvia; Cerquitelli, Tania; Wrembel, Robert; Quercia, Daniele

Abstract: ... Expand

Abstract: Collapse

Topics: mobile system internet of things Twitter MongoDB application programming interface

Methods: network analysis

Atypical Sample Regularizer Autoencoder for Cross-Domain Human Activity Recognition

2021 | Information Systems Frontiers | Citations: 0

Authors: Prabono, Aria Ghora; Yahya, Bernardo Nugroho; Lee, Seok-Lyong

Abstract: The sensor-based human activity recognition (HAR) using machine learning require ... Expand

Abstract: The sensor-based human activity recognition (HAR) using machine learning requires a sufficiently large amount of annotated data to realize an accurate classification model. This requirement stimulates the advancement of the transfer learning research area that minimizes the use of labeled data by transferring knowledge from the existing activity recognition domain. Existing approaches transform the data into a common subspace between domains which theoretically loses information, to begin with. Besides, they are based on the linear projection which is bound to linearity assumption and its limitations. Some recent works have already incorporated nonlinearity to find a latent representation that minimizes domain discrepancy based on an autoencoder that includes statistical distance minimization. However, such approach discovers latent representation for both domains at once, which causes sub-optimal representation because both domains compensate each other’s reconstruction error during the training. We propose an autoencoder-based approach on domain adaptation for sensor-based HAR. The proposed approach learns a latent representation which minimizes the discrepancy between domains by reducing statistical distance. Instead of learning representation of both domains simultaneously, our method is a two-phase approach which first learns the representation for the domain of interest independently to ensure its optimality. Subsequently, the useful information from the existing domain is transferred. We test our approach on the publicly available sensor-based HAR datasets, using cross-domain setup. The experimental result shows that our approach significantly outperforms the existing ones. Collapse

Topics: online learning system smart wearable device

Methods: data transformation autoencoder literature study principal component analysis machine learning

Towards The Measurement Of Business Intelligence Maturity

2013 | European Conference On Information Systems | Citations: 6

Authors: Raber, David; Wortmann, Felix; Winter, Robert

Abstract: For the systematic evolution of interdisciplinary socio-technical systems, such ... Expand

Abstract: For the systematic evolution of interdisciplinary socio-technical systems, such as business intelligence (BI), artifacts are needed that comprehensively address multifaceted challenges. To support these strategic improvement initiatives, we proposed a BI maturity model (MM) in prior research. In this paper, we develop an approach to measure maturity of BI in organizations, thereby operationalizing our existing BI MM. In fact, a new maturity measurement instrument is developed that can be used in empirical research. According to the proposed approach, BI maturity can be calculated on the basis of 25 items and organizations are classified into five maturity levels. An exemplary application, using data from 92 organizations, shows that most organizations reside on maturity level three. Furthermore, we investigate the relationship between BI maturity and business benefits of BI. Findings showed that mature organizations have achieved significantly higher business benefits from BI than organizations on lower maturity levels. Collapse

Topics: organizational value information systems strategy service oriented architecture

Methods: survey partial least squares regression survey design PLS tool qualitative interview

An Instrument for Measuring SOA Maturity

2011 | International Conference on Information Systems | Citations: 4

Authors: Joachim, Nils; Beimborn, Daniel; Weitzel, Tim

Abstract: Existing empirical research on the business value of Service-Oriented Architectu ... Expand

Abstract: Existing empirical research on the business value of Service-Oriented Architectures (SOA) has only measured the extent of SOA adoption – but not maturity – to determine whether typical goals, like increased business agility or IT cost reduction, could be achieved. However, a widely implemented SOA might be less mature than an SOA adopted only in particular areas of the organization, which in turn can lead to mismeasurement and misinterpretation. On the other side, the few existing SOA maturity frameworks that have been specified by previous researchers lack valid operationalizations to make them applicable to empirical research. Ready to use items and scales for evaluating the particular maturity level of an organization are missing. Collapse

Topics: organizational value business process management IT investment systems development IS management

Methods: maturity model survey design science survey design field study

Histogram distance-based Bayesian Network structure learning: A supervised classification specific approach

2009 | Decision Support Systems | Citations: 0

Authors: Sierra, B.; Lazkano, E.; Jauregi, E.; Irigoien, I.

Abstract: In this work we introduce a methodology based on histogram distances for the aut ... Expand

Abstract: In this work we introduce a methodology based on histogram distances for the automatic induction of Bayesian Networks (BN) from a file containing cases and variables related to a supervised classification problem. The main idea consists of learning the Bayesian Network structure for classification purposes taking into account the classification itself, by comparing the class distribution histogram distances obtained by the Bayesian Network after classifying each case. The structure is learned by applying eight different measures or metrics: the Cooper and Herskovits metric for a general Bayesian Network and seven different statistical distances between pairs of histograms. The results obtained confirm the hypothesis of the authors about the convenience of having a BN structure learning method which takes into account the existence of the special variable (the one corresponding to the class) in supervised classification problems. Collapse

Topics: database system missing data

Methods: computational algorithm Bayesian network experiment machine learning chi squared test

Evaluating and Timing Predictive Data Mining Models Using Receiver Operating Characteristic Curves

2004 | Journal of Management Information Systems | Citations: 0

Authors: Sinha, Atish P.; May, Jerrold H.

Abstract: In this study, we conduct an empirical analysis of the performance of five popul ... Expand

Abstract: In this study, we conduct an empirical analysis of the performance of five popular data mining methods--neural networks, logistic regression, linear discriminant analysis, decision trees, and nearest neighbor--on two binary classification problems from the credit evaluation domain. Whereas most studies comparing data mining methods have employed accuracy as a performance measure, we argue that, for problems such as credit evaluation, the focus should be on minimizing misclassification cost. We first generate receiver operating characteristic (ROC) curves for the classifiers and use the area under the curve (AUC) measure to compare aggregate performance of the five methods over the spectrum of decision thresholds. Next, using the ROC results, we propose a method for tuning the classifiers by identifying optimal decision thresholds. We compare the methods based on expected costs across a range of cost-probability ratios. In addition to expected cost and AUC, we evaluate the models on the basis of their generalizability to unseen data, their scalability to other problems in the domain, and their robustness against changes in class distributions. We found that the performance of logistic regression and neural network models was superior under most conditions. In contrast, decision tree and nearest neighbor models yielded higher costs, and were much less generalizable and robust than the other models. An important finding of this research is that the models can be effectively tuned post hoc to make them cost sensitive, even though they were built without incorporating misclassification costs. Collapse

Topics: logistics management information system use

Methods: machine learning decision tree classification artificial neural network logistic regression linear discriminant analysis

Theories: theory of case-based reasoning