Data Science Commons^™

Mining Human Activity Using Dimensionality Reduction And Pattern Recognition, Ismail El Moudden, Mounir Ouzir, Badreddine Benyacoub, Souad El Bernoussi

Research and Infrastructure Service Enterprise (RISE) Faculty Publications

Human activity recognition (HAR) is an emerging research topic in pattern recognition, especially in computer vision. The main objective of human activity recognition is to automatically detect and analyze human activities from the information acquired from different sensors. Human activity prediction using big data remains a challengingly open problem. Several approaches have recently been developed in order to find practical ways to solve high dimensionality of data problems. The aim of this study is to attempt, using data mining techniques, to deal with HAR modeling involving a significant number of variables in order to identify relevant parameters from data and …

An Improved Smote Algorithm Based On Genetic Algorithm For Imbalanced Data Collection, Qiong Gu, Xian-Ming Wang, Zhao Wu, Bing Ning, Chun-Sheng Xin

Electrical & Computer Engineering Faculty Publications

Classification of imbalanced data has been recognized as a crucial problem in machine learning and data mining. In an imbalanced dataset, minority class instances are likely to be misclassified. When the synthetic minority over-sampling technique (SMOTE) is applied in imbalanced dataset classification, the same sampling rate is set for all samples of the minority class in the process of synthesizing new samples, this scenario involves blindness. To overcome this problem, an improved SMOTE algorithm based on genetic algorithm (GA), namely, GASMOTE was proposed. First, GASMOTE set different sampling rates for different minority class samples. A combination of the sampling rates …

Results And Challenges In Visualizing Analytic Provenance Of Text Analysis Tasks Using Interaction Logs, Rhema Linder, Alyssa M. Peña, Sampath Jayarathna, Eric D. Ragan

Computer Science Faculty Publications

After data analysis, recalling and communicating the steps and rationale followed during the analysis can be difficult. This paper explores the use of interaction logs to generate summaries of an analyst's interest based on interactions with specific data items in a text analysis scenario. Our approach uses data-interaction events as a proxy for user interest in and experience of information. Logging can produce verbose logs that detail all available readable content, so the discussed approach uses topic modeling (LDA) over different time segments to summarize the verbose information and generate visualizations of the history of user interest. Our preliminary results …

Engineering Analytics: Research Into The Governance Structure Needed To Integrate The Dominant Design Methodologies, Teddy Steven Cotter

Engineering Management & Systems Engineering Faculty Publications

In the ASEM-IAC 2014, Cotter (2014) explored the current state of engineering design, identified the dominate approaches to engineering design, discussed potential contributions from the new field of data analytics to engineering design, and proposed an Engineering Analytics framework that integrates the dominate engineering design approaches and data analytics within a human-intelligence/machine-intelligence (HI-MI) design architecture. This paper reports research applying ontological engineering to integrate the dominate engineering design methodologies into a systemic engineering design decision governance architecture.

Statistical Engineering: A Causal-Stochastic Modeling Research Update, Teddy Steven Cotter

Engineering Management & Systems Engineering Faculty Publications

In the ASEM-IAC 2012, Cotter (2012) summarized prior works that led to the proposal for statistical engineering, identified the gaps in knowledge that statistical engineering needs to address, explored additional gaps in knowledge not addressed in the prior works, set forth a working definition of and body of knowledge for statistical engineering, and set forth proposals of potential systems contributions the Engineering Management profession could make toward the development of statistical engineering. In 2014, the ASQ Statistics Division, DOT&E, NASA, and IDA co-sponsored a Statistical Engineering Agreement to jointly research development of the discipline of statistical engineering. The statistics community …

Adaptive Graph Construction For Isomap Manifold Learning, Loc Tran, Zezhong Zheng, Guoquing Zhou, Jiang Li, Karen O. Egiazarian (Ed.), Sos S. Agaian (Ed.), Atanas P. Gotchev (Ed.)

Electrical & Computer Engineering Faculty Publications

Isomap is a classical manifold learning approach that preserves geodesic distance of nonlinear data sets. One of the main drawbacks of this method is that it is susceptible to leaking, where a shortcut appears between normally separated portions of a manifold. We propose an adaptive graph construction approach that is based upon the sparsity property of the ℓ₁ norm. The ℓ₁ enhanced graph construction method replaces k-nearest neighbors in the classical approach. The proposed algorithm is first tested on the data sets from the UCI data base repository which showed that the proposed approach performs better than …

Data Completion Methods For Improved Developmental Stage Annotation Of Drosophila Embryos In Images, Chitsanu Janyalikit

Electrical & Computer Engineering Theses & Dissertations

Drosophila melanogaster is a dominant model organism for studying the function of animal genes in initial stages of embryogenesis. Usually, images containing Drosophila gene expression patterns are captured at different developmental stages to study the interconnection of animal genes. To achieve most biologically meaningful results, gene expression images from a similar stage should be compared. Currently, biologists manually classify embryos in images into different stages, which is time intensive and infeasible for current massively produced gene expression images. Therefore, there is a need to develop an automatic system for the annotation.

Gene expression information in embryo images usually appears as …

Assessing Organizational Effectiveness Through The Competing Values Framework A Data Envelopment Approach, Raghavender Macherla

Engineering Management & Systems Engineering Theses & Dissertations

This study proposes a model to diagnose organizations using the mathematical principles of data envelopment analysis (DEA) to the variables generated using competing values framework (CVF) in order to evaluate overall organizational effectiveness. The notion of organizational effectiveness is abstract and difficult to measure due to its complexity and multi-functional nature. Over the years, measurement of organizational effectiveness has remained a challenge due to the lack of agreement on the factors that should be assessed to determine effectiveness. This research is aimed at shedding some light into this topic by using data envelopment analysis as a tool to measure relative …

The Bivariate Erlang And Its Application In Modeling Recurrence Times Of Kidney Dialysis Data, Norou Diawara, S.H. Sathish Indika, Melva Grant, Edgard M. Maboudou-Tchao

Mathematics & Statistics Faculty Publications

Recent advances in computer modeling allows us to find closer fits to data. Our emphasis is on the interdependence between occurrence at kidney dialysis. The interdependence between kidney dialysis occurrences is modelled by a bivariate exponential that we propose in this article. The application is shown on the McGilchrist and Aisbett kidney data set with the use of the exponential distribution. The proposed bivariate exponential model has exponential marginal densities, correlated via a latent random variables and with finite probability of simultaneous occurrence. Extension of the model to a bivariate Erlang type distribution with same shape parameter is presented.

Classification With Hidden Markov Model, Badreddine Benyacoub, Souad Elbernoussi, Abdelhak Zoglat, Ismail El Moudden

Research and Infrastructure Service Enterprise (RISE) Faculty Publications

Classification and statistical learning by hidden markov model has achieved remarkable progress in the past decade. They have been applied in many areas like speech recognition and handwriting recognition. However, learning by Hidden Markov Model (HMM) is still restricted to supervised problems. In this paper, we propose a new learning method based on HMM techniques estimations, to built a model for classification. The approach consists of evaluation of the probability to belonging in one group, given the observations by a linear classifier. Our developed algorithm is based on discrete states and discrete observations cases of HMM. Experimental results show that …

Hyperspectral Image Classification Using A Spectral-Spatial Sparse Coding Model, Ender Oguslu, Guoqing Zhou, Jiang Li, Lorenzo Bruzzone (Ed.)

Electrical & Computer Engineering Faculty Publications

We present a sparse coding based spectral-spatial classification model for hyperspectral image (HSI) datasets. The proposed method consists of an efficient sparse coding method in which the l₁/l_q regularized multi-class logistic regression technique was utilized to achieve a compact representation of hyperspectral image pixels for land cover classification. We applied the proposed algorithm to a HSI dataset collected at the Kennedy Space Center and compared our algorithm to a recently proposed method, Gaussian process maximum likelihood (GP-ML) classifier. Experimental results show that the proposed method can achieve significantly better performances than the GP-ML classifier when training data …

Model Individualization For Real-Time Operator Functional State Assessment, Guangfan Zhang, Roger Xu, Wei Wang, Aaron A. Pepe, Feng Li, Jiang Li, Frederick Mckenzie, Tom Schnell, Nick Anderson, Dean Heitkamp

Electrical & Computer Engineering Faculty Publications

Proper assessment of Operator Functional State (OFS) and appropriate workload modulation offer the potential to improve mission effectiveness and aviation safety in both overload and under-load conditions. Although a wide range of research has been devoted to building OFS assessment models, most of the models are based on group statistics and little or no research has been directed towards model individualization, i.e., tuning the group statistics based model for individual pilots. Moreover, little emphasis has been placed on monitoring whether the pilot is disengaged during low workload conditions. The primary focus of this research is to provide a real-time engagement …

Real-Time Anomaly Detection In Full Motion Video, Glenn Konowicz,, Jiang Li, Donnie Self (Ed.)

Electrical & Computer Engineering Faculty Publications

Improvement in sensor technology such as charge-coupled devices (CCD) as well as constant incremental improvements in storage space has enabled the recording and storage of video more prevalent and lower cost than ever before. However, the improvements in the ability to capture and store a wide array of video have required additional manpower to translate these raw data sources into useful information. We propose an algorithm for automatically detecting anomalous movement patterns within full motion video thus reducing the amount of human intervention required to make use of these new data sources. The proposed algorithm tracks all of the objects …

Sensitivity Analysis Framework For Large And Complex Simulation Models, Ghaith Rabadi, Shannon Bowling, Charles Keating, Resit Unal

Engineering Management & Systems Engineering Faculty Publications

In this paper, a framework for conducting Sensitivity Analysis (SA) on large and complex simulation models is introduced. The framework consists of components that are designed to make the SA a systematic process that is easy to manage and follow by simulation analysts and practitioners. Unlike local SA (one-variable-at-a-time SA), the method presented here is variance-based and it is rooted in the field of Design of Experiments (DoE) where Input Variables are varied and Output Variables are measured. Based on the DoE results, a risk scoring system is developed to identify the sensitivity of the Input Variables, and as a …

Enhancing The Collection Process For The Delphi Technique, Petros Katsioloudis, John Brocato (Ed.)

STEMPS Faculty Publications

The purpose of this manuscript is to describe a process that enhances the data collection process for a Delphi technique. The approach consists of online platforms to expedite the process and reinforce the validity of the Delphi technique. The context of the study was the identification of quality indicators for visual-based learning material development for Technology Education programs for grades 7-12.

Object Reuse And Exchange, Michael L. Nelson, Carl Lagoze, Herbert Van De Sompel, Pete Johnston, Robert Sanderson, Simeon Warner, Jürgen Sieck (Ed.), Michael A. Herzog (Ed.)

Computer Science Faculty Publications

The Open Archives Object Reuse and Exchange (OAI-ORE) project defines standards for the description and exchange of aggregations of Web resources. The OAI-ORE abstract data model is conformant with the Architecture of the World Wide Web and leverages concepts from the Semantic Web, including RDF descriptions and Linked Data. In this paper we provide a brief review of a motivating example and its serialization in Atom.

Distributed Cluster-Based Outlier Detection In Wireless Sensor Networks, Swetha Gali

Electrical & Computer Engineering Theses & Dissertations

Wireless sensor networks find several potential applications in a variety of fields, such as environmental monitoring and control, battlefields, surveillance, smart buildings, human health monitoring, etc. These sensor networks consist of a large number of very tiny, inexpensive, and low power sensor nodes, which are deployed in a variety of harsh environments that may result in the sensor data getting corrupted. It is thus critical to detect and report these abnormal values in the sensor data, in order to have a better understanding of the monitored environment. Detection of the abnormal values is of special interest for the sensor network …

Exploration Of Computational Methods For Classification Of Movement Intention During Human Voluntary Movement From Single Trial Eeg, Ou Bai, Peter Lin, Sherry Vorbach, Jiang Li, Steve Furlani, Mark Hallett

Electrical & Computer Engineering Faculty Publications

Objective: To explore effective combinations of computational methods for the prediction of movement intention preceding the production of self-paced right and left hand movements from single trial scalp electroencephalogram (EEG).

Methods: Twelve naïve subjects performed self-paced movements consisting of three key strokes with either hand. EEG was recorded from 128 channels. The exploration was performed offline on single trial EEG data. We proposed that a successful computational procedure for classification would consist of spatial filtering, temporal filtering, feature selection, and pattern classification. A systematic investigation was performed with combinations of spatial filtering using principal component analysis (PCA), independent component analysis …

Dimensionality Reduction Using Non-Linear Principal Components Analysis, Tara Singh

Electrical & Computer Engineering Theses & Dissertations

Advances in data collection and storage capabilities during the past decades have led to an information overload in most sciences. Traditional statistical methods break down partly because of the increase in the number of observations, but mostly because of the increase in the number of variables associated with each observation. While certain methods can construct predictive models with high accuracy from high-dimensional data, it is still of interest in many applications to reduce the dimension of the original data prior to any modeling of the data. Patterns in the data can be hard to find in data of high dimensionality, …

Information Visualization Methods And Techniques Using Web Services, Kevin Dupigny

Computational Modeling & Simulation Engineering Theses & Dissertations

Data, even in small amounts, can be difficult to interpret quickly and efficiently. Data domains are continually expanding and making this data available and usable in collaborative data environment is a challenge that was not addressed by legacy systems. Such a framework would make the existence of a global information sharing system feasible. Visualizing this complex dataset presents a complex task for current static interface solutions. A flexible layered service oriented model is presented in this work to approach this challenge. The visual scope of human perception is largely untapped by today's data visualization applications. Human mental processes are adept …

A Component Based Formal Ontology Model, A Method For Evaluating Such A Model, And The Results Of An Application Of That Method, Charles Tumitsa

Computational Modeling & Simulation Engineering Theses & Dissertations

For systems to truly communicate effectively, more than just data interchange is necessary. The systems must achieve conceptual interoperability, which implies communicating with understanding.

The LCIM (Levels of Conceptual Interoperability Model) describes a number of levels of interoperability, each of which shows systems with an increased amount of conceptual understanding of the other systems they are communicating with. The lowest levels of this model, technical and syntactic interoperability, are achieved (and there are many existing examples) via data exchange in a number of different ways. Research has shown, however, that to attain higher levels of conceptual understanding, that there must …

Synchronization And Multiple Group Server Support For Kepler, K. Maly, M. Zubair, H. Siripuram, S. Zunjarwad, Yannis Manolopoulos (Ed.), Joaquim Filipe (Ed.), Panos Constantopoulos (Ed.), José Cordeiro (Ed.)

Computer Science Faculty Publications

In the last decade literally thousands of digital libraries have emerged but one of the biggest obstacles for dissemination of information to a user community is that many digital libraries use different, proprietary technologies that inhibit interoperability. Kepler framework addresses interoperability and gives publication control to individual publishers. In Kepler, OAI-PMH is used to support "personal data providers" or "archivelets".". In our vision, individual publishers can be integrated with an institutional repository like Dspace by means of a Kepler Group Digital Library (GDL). The GDL aggregates metadata and full text from archivelets and can act as an OAI-compliant data provider …

Recommender Systems For Multimedia Libraries: An Evaluation Of Different Models For Datamining Usage Data, Raquel Oliveira Araujo

Computer Science Theses & Dissertations

Many recommender systems exist today to help users deal with the large growth in the amount of information available in the Internet. Most of these recommender systems use collaborative filtering or content-based techniques to present new material that would be of interest to a user. While these methods have proven to be effective, they have not been designed specifically for multimedia collections. In this study we present a new method to find recommendations that is not dependent on traditional Information Retrieval (IR) methods and compare it to algorithms that do rely on traditional IR methods. We evaluated these algorithms using …