Analytics Methodology Research Articles

Leveraging Predictive Analytics to Control and Coordinate Operations, Asset Loading, and Maintenance

This paper aims to advance decision making in power systems by proposing an integrated framework that combines sensor data analytics and optimization. Our modeling framework consists of two components: first, a predictive analytics methodology that uses real-time sensor data to predict future degradation and remaining lifetime of generators as a function of the loading conditions, and second, a mixed integer optimization model that transforms these predictions into cost-optimal maintenance and operational decisions. We model the key balance between meeting demand with very high confidence and at the same time prolonging the lifetime of generation assets. To do so, we encapsulate stochastic loading-dependent predictive models for asset condition within our optimization model. The methodology is validated and evaluated using IEEE 118-bus system that has been augmented using real-world sensor-based vibration signals from rotating machinery to emulate physical degradation of generators. Our experiments suggest that the proposed framework provides considerable improvements over conventional methods in terms of cost and reliability.

A Learning Analytics Methodology for Understanding Social Interactions in MOOCs

One of the characteristics of massive open online courses (MOOCs) is that the overall number of social interactions tend to be higher than in traditional courses, hindering the analysis of social learning. Learners typically ask or answer questions using the forum. This makes messages a rich source of information, which can be used to infer learners' behavior and outcomes. It is not feasible for teachers to process all forum messages and automated tools and analysis are required. Although there are some tools for analyzing learners' interactions, there is a need for methodologies and integrated tools that help to interpret the learning process based on social interactions in the forum. This paper presents the 3S (Social, Sentiments, Skills) learning analytics methodology for analyzing forum interactions in MOOCs. This methodology considers a temporal analysis combining the social, sentiments, and skill dimensions that can be extracted from forum data. We also introduce LAT∃S, a learning analytics tool for edX/Open edX related to the three dimensions (3S), which includes visualizations to guide the proposed methodology. We apply the 3S methodology and the tool to an MOOC on Java programming. Results showed, among others, the action-reaction effect produced when learners increase their participation after instructor's events. Moreover, a decrease of positive sentiments over time and before deadlines of open-ended assignments was also observed and that there were certain skills, which caused more troubles (e.g., arrays and loops). These results acknowledge the importance of using a learning analytics methodology to detect problems in MOOCs.

Visual exploration of mobility dynamics based on multi-source mobility datasets and POI information

Learning more about human mobility is crucial for official decision makers and urban planners. Mobility datasets characterize human daily travel behaviors. Most current researches only studied human dynamics from one kind of mobility dataset. However, people may use different means of transportation to different places for different purposes. Besides, the spatial distributions of different types of point of interests (POIs) reflect the land-use types. How to jointly analyze the multi-source mobility datasets and POI information is a great challenge. In this paper, we adopt multi-source datasets, including taxi dataset, public bicycle system dataset and POI dataset, and propose a visual analytics methodology to explore human mobility dynamics insightfully. Two region–feature–time tensors are constructed first, and a tensor decomposition method is employed to classify the mobility patterns automatically. Then, a new POI–mobility glyph is designed to visualize multi-source datasets in a compact manner. Several interactive visual views are also designed to visualize the spatiotemporal patterns from global, regional and locational perspectives. Case studies based on real-world datasets demonstrate the effectiveness of our method, which supports the visual reasoning of trip purposes and mixed urban functions.

Audit Data Analytics Research—An Application of Design Science Methodology

SYNOPSIS This introduction to Audit Data Analytics Research overviews the forum's five articles that showcase recent advances in audit data analytics technology and methodology. The articles are discussed through the prism of design science research that originates in engineering and computer science. In contrast with natural and social sciences that aim to develop and test theories about the world, the objective of design science is to create new artifacts that are useful for solving important practical problems. In audit research, design science methodology was originally used implicitly in early studies devoted to developing and evaluating audit analytical procedures and audit sampling techniques. The recent advances in information technology necessitate renewed attention to this research methodology especially given the profound changes in accounting, auditing, and business processes currently underway.

Rapid Transit Systems: Smarter Urban Planning Using Big Data, In-Memory Computing, Deep Learning, and GPUs

Rapid transit systems or metros are a popular choice for high-capacity public transport in urban areas due to several advantages including safety, dependability, speed, cost, and lower risk of accidents. Existing studies on metros have not considered appropriate holistic urban transport models and integrated use of cutting-edge technologies. This paper proposes a comprehensive approach toward large-scale and faster prediction of metro system characteristics by employing the integration of four leading-edge technologies: big data, deep learning, in-memory computing, and Graphics Processing Units (GPUs). Using London Metro as a case study, and the Rolling Origin and Destination Survey (RODS) (real) dataset, we predict the number of passengers for six time intervals (a) using various access transport modes to reach the train stations (buses, walking, etc.); (b) using various egress modes to travel from the metro station to their next points of interest (PoIs); (c) traveling between different origin-destination (OD) pairs of stations; and (d) against the distance between the OD stations. The prediction allows better spatiotemporal planning of the whole urban transport system, including the metro subsystem, and its various access and egress modes. The paper contributes novel deep learning models, algorithms, implementation, analytics methodology, and software tool for analysis of metro systems.

Smarter Traffic Prediction Using Big Data, In-Memory Computing, Deep Learning and GPUs.

Road transportation is the backbone of modern economies, albeit it annually costs million deaths and trillions of dollars to the global economy, and damages public health and the environment. Deep learning is among the leading-edge methods used for transportation-related predictions, however, the existing works are in their infancy, and fall short in multiple respects, including the use of datasets with limited sizes and scopes, and insufficient depth of the deep learning studies. This paper provides a novel and comprehensive approach toward large-scale, faster, and real-time traffic prediction by bringing four complementary cutting-edge technologies together: big data, deep learning, in-memory computing, and Graphics Processing Units (GPUs). We trained deep networks using over 11 years of data provided by the California Department of Transportation (Caltrans), the largest dataset that has been used in deep learning studies. Several combinations of the input attributes of the data along with various network configurations of the deep learning models were investigated for training and prediction purposes. The use of the pre-trained model for real-time prediction was explored. The paper contributes novel deep learning models, algorithms, implementation, analytics methodology, and software tool for smart cities, big data, high performance computing, and their convergence.

Exploring the ethical implications of business analytics with a business ethics canvas

The ethical aspects of data science and artificial intelligence have become a major issue. Organisations that deploy data scientists and operational researchers (OR) must address the ethical implications of their use of data and algorithms. We review the OR and data science literature on ethics and find that this work is pitched at the level of guiding principles and frameworks and fails to provide a practical and grounded approach that can be used by practitioners as part of the analytics development process. Further, given the advent of the General Data Protection Regulation (GDPR) an ethical dimension is likely to become an increasingly important aspect of analytics development. Drawing on the business analytics methodology (BAM) developed by Hindle and Vidgen (2018) we tackle this challenge through action research with a pseudonymous online travel company, EuroTravel. The method that emerges uses an opportunity canvas and a business ethics canvas to explore value creation and ethical aspects jointly. The business ethics canvas draws on the Markkula Center's five ethical principles (utility, rights, justice, common good, and virtue) to which explicit consideration of stakeholders is added. A contribution of the paper is to show how an ethical dimension can be embedded in the everyday exploration of analytics development opportunities, as distinct from a stand-alone ethical decision-making tool or as an overlay of a general set of guiding principles. We also propose that value and ethics should not be viewed as separate entities, rather they should be seen as inseparable and intertwined.

Knowledge Generation Using Sentiment Classification Involving Machine Learning on E-Commerce

Sentiment analysis manages the computational treatment of conclusion, notion, and content subjectivity. In this article, three sentiment classes such as positive, negative and neutral emotions have been demonstrated by appropriate features from raw unstructured data followed by data preprocessing steps. Applying best in class social analytics methodology to examine the sentiments embedded with purchaser remarks, encourages both producer and individual customers. Machine learning methods such as Naïve Bayes, maximum entropy classification, Deep Neural Networks were used upon the data, extracted from some websites such as Samsung and Apple for sentiment classification. In the online business arena, the application of sentiment classification explores a great opportunity. The subsidy of such an investigation is that associations can apply the proposed social examination framework to exploit the entire social information on the web and therefore improve their proper blueprint promoting strategies corresponding business.

Weapon Combat Effectiveness Analytics Using Big Data and Simulations: A Literature Review

<div class="section abstract"><div class="htmlview paragraph">The Weapon Combat Effectiveness (WCE) analytics is very expensive, time-consuming, and dangerous in the real world because we have to create data from the real operations with a lot of people and weapons in the actual environment. The Modeling and Simulation (M&S) of many techniques are used for overcoming these limitations. Although the era of big data has emerged and achieved a great deal of success in a variety of fields, most of WCE research using the Defense Modeling and Simulation (DM&S) techniques studied have considered a lot of assumptions and limited scenarios without the help of big data technologies. Furthermore, WCE analytics using previous methodologies cannot help but get the bias results. This paper reviews and combines the basic knowledge for the new WCE analytics methodology using big data and M&S to overcome these problems of bias. Then this paper reviews the general overview of WCE, DM&S, and big data. Also, it establishes WCE classification and identifies the research challenges and solutions. The new WCE analytics methodology is discussed based on the research solutions. This paper will have a role to further serve as the stepping-stone of the future research for the interested researchers.</div></div>

Materials data analytics for 9% Cr family steel

A materials data analytics (MDA) methodology was developed in this study to evaluate publicly available information on 9% Cr family steel and to handle nonlinear relationships and the sparsity in materials data for this alloy class. The overarching goal is to accelerate the design process as well as to reduce the time and expense associated with qualification testing of new alloys for fossil energy applications. Data entries in the analyzed data set for 82 iron‐base alloy compositions, several processing parameters, and results of tensile mechanical tests selected for this study were arranged in 34 columns by 915 rows. While detailed microstructural information was not available, it is assumed that the compositional space for the 9 to 12% Cr steels is limited such that all data entries have a tempered martensitic microstructure during service. Establishing a hierarchy of first‐order trends in the publicly available data requires the MDA to filter out the biases. Complexity of the phase transformations and microstructure evolution in the multicomponent alloys (using 21 chemical elements) with major influence on mechanical properties, leads to inefficiency in direct application of unbiased linear regression across the entire data space. To address the nonlinearity, analyses of tensile data were performed in composition‐based clusters. Clusters corresponding to moderately frequent patterns and maximized information gain were further refined by using p‐norm distance measures, matching the alloy classification groups adopted by industry. The evolutionary method of propagating an ensemble of competing cluster‐based models proved to be a viable option in dealing with scarce, multidimensional data.

Predictive analytics methodology for smart qualification testing of electronic components

In electronics manufacturing, the required quality of electronic modules (e.g. packaged electronic devices) are evaluated through qualification testing using standards and user-defined requirements. The challenge for the electronics industry is that product qualification testing is time-consuming and costly. This paper focuses on the development and demonstration of a novel approach for smarter qualification using test data from the production line along with integrated computational techniques for data mining/analytics and data-driven forecasting (i.e. prognostics) modelling. The most common type of testing in the electronics industry—sequentially run electrical multi-parameter tests on the Device-under-Test (DUT), is considered. The proposed data mining (DM) framework can identify the tests that have strong correlation to pending failure of the device in the qualification (tests sensitive to pending failure) as well as to evaluate the similarity in test measurements, thus generating knowledge on potentially redundant tests. Mining the data in this context and with the proposed approach represents a major new contribution because it uncovers embedded knowledge and information in the production test data that can enable intelligent optimisation of the tests’ sequence and reduce the number of tests. The intelligent manufacturing concept behind the development of data-driven prognostics models using machine learning techniques is to use data only from a small number of tests from the full qualification specification as training data in the process of model construction. This model can then forecast the overall qualification outcome for a DUT—Pass or Fail—without performing all other remaining tests. The novelty in the context of machine learning is in the selection of the data features for the training dataset using results from tests sensitive to pending failure. Support Vector Machine (SVM) binary classifiers SVM models built with data from tests sensitive to the outcome that the module will fail are shown to have superior performance compared with models trained with other datasets of tests. Case studies based on the use of real industrial production test data for an electronic module are included in the paper to demonstrate and validate the computational approach. This work is both novel and original because at present, to the best knowledge of the authors, such predictive analytics methodology applied to qualification testing and providing benefits of test time and hence cost reduction are non-existent in the electronics industry. The integrated data analytics-prognostics approach, deployable for both off-line and in-line optimisation of production test procedures, has the potential to transform current practices by exploiting in a smarter way information and knowledge available with large datasets of qualification test data.

Examining Network Segmentation for Traffic Safety Analysis With Data-Driven Spectral Analysis

Network segmentation is foundational and critical to traffic safety analysis. Existing approaches to conduct segmentation require engineering judgement and are subject to a lack of standard metrics for assessing segmentation performance. This paper presents a novel methodology for data-driven analytics of crash distribution, crash aggregation, and network segmentation. It provides general solutions to determine optimal segment lengths for rigorous safety analysis, and extends the knowledge of crash distribution and aggregation for innovating segment-based safety analysis. The methodology is based on a redesigned spectral analysis of crash density in the spatial frequency domain (SFD) in which frequency components represent the natural patterns how crashes occur along roadways. By proposing the one-dimensional spatial frequency domain analysis (SFDA), this paper reveals the characteristic of power spectral concentration within the low frequency band for crash distribution. Based on this finding, this paper further proposes the power spectral segment length (PSSL) for determining optimal segment lengths and the power spectral percentage (PSP) for assessing the segmentation performance. Based on those new concepts and inferences, the paper proposes the low-pass filtering (LPF) method that outperforms the sliding window (SW) method, and the improved wavelet-based method that identifies high-risk segments properly. Those new techniques are easy to implement and ready for practical application. This research illustrates that interdisciplinary and innovative analytics combined with high-quality data collected by intelligent transportation infrastructure can reshape the fundamental knowledge and conventional paradigms in traffic safety.

Exploring Evolution of Dynamic Networks via Diachronic Node Embeddings.

Dynamic networks evolve with their structures changing over time. It is still a challenging problem to efficiently explore the evolution of dynamic networks in terms of both their structural and temporal properties. In this paper, we propose a visual analytics methodology to interactively explore the temporal evolution of dynamic networks in the context of their structure. A novel diachronic node embedding method is first proposed to learn latent representations of the structural and temporal features of nodes in a vector space. Diachronic node embeddings are then used to discover communities with similar structural proximity and temporal evolution patterns. A visual analytics system is designed to enable users to visually explore the evolutions of nodes, communities, and the network as a whole in terms of their structural and temporal properties. We evaluate the effectiveness of our method using artificial and real-world dynamic networks and comparisons with previous methods.

Visual analytics of cellular signaling data

Cellular signaling data is a type of traffic data, which contains rich spatio-temporal information. Rather than studying the trajectories of individuals, we propose a visual analytics methodology to analyze the crowd flows among a geographical network extracted from real-time cellular signaling data. We design a suite of visualization techniques to explore and reveal mobility patterns over the networks of spatiotemporal clustering. The feasibility of our approach was verified on a real real-time cellular signaling dataset in one week.

Application of latent semantic analysis in AIS academic research

This study provides insights about the historical, intellectual structure, and trends of academic research themes in journals specifically dedicated to Accounting Information Systems (AIS) research—International Journal of Accounting Information Systems (IJAIS), its predecessor journal, Advances in Accounting Information Systems (AiAIS), and Journal of Information Systems (JIS). Using Latent Semantic Analysis, a statistical text analytics methodology that can uncover the conceptual content within unstructured data, this study identifies 14 prevalent academic research themes in AiAIS, IJAIS, and JIS from 1986 to 2015 and provides graphs that visualize thematic trends over time, including by journal. Certain themes have remained consistent in their study over the timeframe, while others have increased or diminished. Certain themes have matured while others appear to still be maturing at the end of the timeframe. Thematic trends by source journal suggest that no journal has dominated publishing specific significant themes in AIS academic research.

Parallel Aspect‐Oriented Sentiment Analysis for Sales Forecasting with Big Data

While much research work has been devoted to supply chain management and demand forecast, research on designing big data analytics methodologies to enhance sales forecasting is seldom reported in existing literature. The big data of consumer‐contributed product comments on online social media provide management with unprecedented opportunities to leverage collective consumer intelligence for enhancing supply chain management in general and sales forecasting in particular. The main contributions of our work presented in this study are as follows: (1) the design of a novel big data analytics methodology that is underpinned by a parallel aspect‐oriented sentiment analysis algorithm for mining consumer intelligence from a huge number of online product comments; (2) the design and the large‐scale empirical test of a sentiment enhanced sales forecasting method that is empowered by a parallel co‐evolutionary extreme learning machine. Based on real‐world big datasets, our experimental results confirm that consumer sentiments mined from big data can improve the accuracy of sales forecasting across predictive models and datasets. The managerial implication of our work is that firms can apply the proposed big data analytics methodology to enhance sales forecasting performance. Thereby, the problem of under/over‐stocking is alleviated and customer satisfaction is improved.

Using Web Analytics to Assess Traffic to the Mandela Portal: The Case of African Countries

ABSTRACTThis article is a longitudinal assessment of the visitors to the Mandela Portal between 2009 and 2016. It is based on web analytics methodology that consists of a four-step process with the first and second steps using data collected by Google Analytics tool. The article outlines trends of global figures as well as specific trends within the top ten African countries. It further explored statistics from the top five African countries: South Africa, Kenya, Nigeria, Tanzania, and Ghana. There are two main findings: those related to web traffic trends and those exploring the trustworthiness of Google Analytics data.

Understanding the Impact of Immersion and Authenticity on Satisfaction Behavior in Learning Analytics Tasks

As business analytics (BA) applications permeate across various industry sectors, the workforce needs to be trained and upskilled to meet the challenges of understanding and implementing analytics methodologies. To achieve payoffs from the resource investments in BA training, it is critical for enterprises to understand an individual’s learning behavior along with the process and outcome-centric satisfaction associated with a collaborative analytics training task. This study focuses on identifying the factors that influence the process of learning during BA training to entry-level BA users. Drawing on the theories of situated cognition, goal setting, and flow, we propose a model that explains how trainees in a group learn through a process that is influenced by the characteristics of BA training context through context authenticity, the traits of trainees through task motivation and preference towards teamwork. Using an experimental design built on data collection and a unique task of real visits to a historic cemetery, we found that context authenticity and task motivation have significant impact on focused immersion, which in turn significantly impacts process and outcome satisfaction for learning an analytics task. Results of this study extend and validate the theories of situated cognition, goal setting, and flow within the context of business analytics training. Based on these findings, we provide recommendations for practitioners for designing effective analytics tasks for better training outcomes.

A Decision Support System to Ease Operator Overload in Multibeam Passive Sonar

Creating human-informative signal processing systems for the underwater acoustic environment that do not generate operator cognitive saturation and overload is a major challenge. To alleviate cognitive operator overload, we present a visual analytics methodology in which multiple beam-formed sonar returns are mapped to an optimized 2-D visual representation, which preserves the relevant data structure. This representation alerts the operator as to which beams are likely to contain anomalous information by modeling a latent distribution of information for each beam. Sonar operators therefore focus their attention only on the surprising events. In addition to the principled visualization of high-dimensional uncertain data, the system quantifies anomalous information using a Fisher Information measure. Central to this process is the novel use of both signal and noise observation modeling to characterize the sensor information. A demonstration of detecting exceptionally low signal-to-noise ratio targets embedded in real-world 33-beam passive sonar data is presented.

MCLEAN: Multilevel Clustering Exploration As Network.

Finding useful patterns in datasets has attracted considerable interest in the field of visual analytics. One of the most common tasks is the identification and representation of clusters. However, this is non-trivial in heterogeneous datasets since the data needs to be analyzed from different perspectives. Indeed, highly variable patterns may mask underlying trends in the dataset. Dendrograms are graphical representations resulting from agglomerative hierarchical clustering and provide a framework for viewing the clustering at different levels of detail. However, dendrograms become cluttered when the dataset gets large, and the single cut of the dendrogram to demarcate different clusters can be insufficient in heterogeneous datasets. In this work, we propose a visual analytics methodology called MCLEAN that offers a general approach for guiding the user through the exploration and detection of clusters. Powered by a graph-based transformation of the relational data, it supports a scalable environment for representation of heterogeneous datasets by changing the spatialization. We thereby combine multilevel representations of the clustered dataset with community finding algorithms. Our approach entails displaying the results of the heuristics to users, providing a setting from which to start the exploration and data analysis. To evaluate our proposed approach, we conduct a qualitative user study, where participants are asked to explore a heterogeneous dataset, comparing the results obtained by MCLEAN with the dendrogram. These qualitative results reveal that MCLEAN is an effective way of aiding users in the detection of clusters in heterogeneous datasets. The proposed methodology is implemented in an R package available at https://bitbucket.org/vda-lab/mclean.