Computer Sciences Commons^™

Active Learning With Efficient Feature Weighting Methods For Improving Data Quality And Classification Accuracy, Justin Martineau, Lu Chen, Doreen Cheng, Amit P. Sheth

Kno.e.sis Publications

Many machine learning datasets are noisy with a substantial number of mislabeled instances. This noise yields sub-optimal classification performance. In this paper we study a large, low quality annotated dataset, created quickly and cheaply using Amazon Mechanical Turk to crowdsource annotations. We describe computationally cheap feature weighting techniques and a novel non-linear distribution spreading algorithm that can be used to iteratively and interactively correcting mislabeled instances to significantly improve annotation quality at low cost. Eight different emotion extraction experiments on Twitter data demonstrate that our approach is just as effective as more computationally expensive techniques. Our techniques save a considerable …

Semantic Modelling Of Smart City Data, Stefan Bischof, Athanasios Karapantelakis, Cosmin-Septimiu Nechifor, Amit P. Sheth, Alessandra Mileo, Payam Barnaghi

Kno.e.sis Publications

Cities present an opportunity for rendering Web of Things-enabled services. According to the World Health Organization, population in cities will double by the middle of this century, while cities deal with increasingly pressing issues such as environmental sustainability, economic growth and citizen mobility. In this paper, we propose a discussion around the need for common semantic descriptions for smart city data to facilitate future services in "smart cities". We present examples of data that can be collected from cities, discuss issues around this data and put forward some preliminary thoughts for creating a semantic description model to describe and help …

Semantics-Enhanced Geoscience Interoperability, Analytics, And Applications, Krishnaprasad Thirunarayan, Amit P. Sheth

Kno.e.sis Publications

We present our research ideas for developing cyberinfrastructure for Geoscience applications developed in the context of the EarthCube initiative, and our NSF-sponsored work on incorporating spatial-temporal-thematic semantics for enhanced querying and feature extraction from sensor data streams.

Visualization Support For Cognitive Sciences, Matt J. Marangoni, Thomas Wischgoll, Yue Zhou, Leslie M. Blaha, Ross Smith, Rhonda J. Vickery

Computer Science and Engineering Faculty Publications

The science of computer graphics and visualization is intertwined in many ways with Cognitive Sciences. On the one hand, computer graphics can lead to virtual environments in which a person is exposed to a virtual scenario. Typically, 3D-capable display technology combined with tracking systems, which are capable of identifying where the person is located at, are deployed to achieve maximal immersion in that the persons point of view is recreated in the virtual scenario. As a result, an impressive experience is created such that that person is navigating the virtual scenario as if it was real. On the other hand, …

With Whom To Coordinate, Why And How In Ad-Hoc Social Media Communications During Crisis Response, Hemant Purohit, Shreyansh Bhatt, Andrew Hampton, Valerie L. Shalin, Amit P. Sheth, John M. Flach

Kno.e.sis Publications

During crises affected people, well-wishers, and observers join social media communities to discuss the event. They often share useful information relevant to response coordination, for example, specific resource needs. However, responders face the challenge of massive data overload and lack the time to monitor social media traffic for important information. Analysis shows that only a small number of event related conversations are actionable. Moreover, responders do not know which sources are trustworthy. To address these challenges, response teams may apply manual filtering methods, resulting in limited coverage and quality. We propose a framework and interface for extracting specific resource-related information …

Mining Contrast Subspaces, Lei Duan, Guanting Tang, Jian Pei, James Bailey, Guozhu Dong, Akiko Campbell, Changjie Tang

Kno.e.sis Publications

In this paper, we tackle a novel problem of mining contrast subspaces. Given a set of multidimensional objects in two classes C₊  and C₋ and a query object o, we want to find top-k subspaces S that maximize the ratio of likelihood of o in C₊  against that in C₋. We demonstrate that this problem has important applications, and at the same time, is very challenging. It even does not allow polynomial time approximation. We present CSMiner, a mining method with various pruning techniques. CSMiner is substantially faster than the baseline method. Our …

Leveraging Social Media And Web Of Data For Crisis Response Coordination, Carlos Castillo, Fernando Diaz, Hemant Purohit

Kno.e.sis Publications

There is an ever increasing number of users in social media (1B+ Facebook users, 500M+ Twitter users) and ubiquitous mobile access (6B+ mobile phone subscribers) who share their observations and opinions. In addition, the Web of Data and existing knowledge bases keep on growing at a rapid pace. In this scenario, we have unprecedented opportunities to improve crisis response by extracting social signals, creating spatio-temporal mappings, performing analytics on social and Web of Data, and supporting a variety of applications. Such applications can help provide situational awareness during an emergency, improve preparedness, and assist during the rebuilding/recovery phase of a …

Hierarchical Interest Graph From Tweets, Pavan Kapanipathi, Prateek Jain, Chitra Venkataramani, Amit P. Sheth

Kno.e.sis Publications

Industry and researchers have identified numerous ways to monetize microblogs for personalization and recommendation. A common challenge across these different works is the identification of user interests. Although techniques have been developed to address this challenge, a flexible approach that spans multiple levels of granularity in user interests has not been forthcoming. In this work, we focus on exploiting hierarchical semantics of concepts to infer richer user interests expressed as a Hierarchical Interest Graph. To create such graphs, we utilize users' tweets to first ground potential user interests to structured background knowledge such as Wikipedia Category Graph. We then adapt …

Comparative Trust Management With Applications: Bayesian Approaches Emphasis, Krishnaprasad Thirunarayan, Pramod Anantharam, Cory Andrew Henson, Amit P. Sheth

Kno.e.sis Publications

Trust relationships occur naturally in many diverse contexts such as collaborative systems, e-commerce, interpersonal interactions, social networks, and semantic sensor web. As agents providing content and services become increasingly removed from the agents that consume them, the issue of robust trust inference and update becomes critical. There is a need to find online substitutes for traditional (direct or face-to-face) cues to derive measures of trust, and create efficient and robust systems for managing trust in order to support decision-making. Unfortunately, there is neither a universal notion of trust that is applicable to all domains nor a clear explication of its …

Cursing In English On Twitter, Wenbo Wang, Lu Chen, Krishnaprasad Thirunarayan, Amit P. Sheth

Kno.e.sis Publications

Cursing is not uncommon during conversations in the physical world: 0.5% to 0.7% of all the words we speak are curse words, given that 1% of all the words are first-person plural pronouns (e.g., we, us, our). On social media, people can instantly chat with friends without face-to-face interaction, usually in a more public fashion and broadly disseminated through highly connected social network. Will these distinctive features of social media lead to a change in people's cursing behavior? In this paper, we examine the characteristics of cursing activity on a popular social media platform - Twitter, involving the analysis of …

Ivus Validation Of Patient Coronary Artery Lumen Area Obtained From Ct Images, Tong Luo, Thomas Wischgoll, Bon Kwon Koo, Yunlong Huo, Ghassan S. Kassab

Computer Science and Engineering Faculty Publications

Aims

Accurate computed tomography (CT)-based reconstruction of coronary morphometry (diameters, length, bifurcation angles) is important for construction of patient-specific models to aid diagnosis and therapy. The objective of this study is to validate the accuracy of patient coronary artery lumen area obtained from CT images based on intravascular ultrasound (IVUS).

Methods and Results

Morphometric data of 5 patient CT scans with 11 arteries from IVUS were reconstructed including the lumen cross sectional area (CSA), diameter and length. The volumetric data from CT images were analyzed at sub-pixel accuracy to obtain accurate vessel center lines and CSA. A new center line …

An Ontology Pattern For Oceanographic Cruises: Towards An Oceanographer's Dream Of Integrated Knowledge Discovery, Adila Krisnadhi, Robert Arko, Suzanne Carbotte, Cynthia Chandler, Michelle Cheatham, Timothy Finin, Pascal Hitzler, Krzysztof Janowicz, Thomas Narock, Lisa Raymond, Adam Shepherd, Peter Wiebe

Computer Science and Engineering Faculty Publications

EarthCube is a major effort of the National Science Foundation to establish a next-generation knowledge architecture for the broader geosciences. Data storage, retrieval, access, and reuse are central parts of this new effort. Currently, EarthCube is organized around several building blocks and research coordination networks. OceanLink is a semantics enabled building block that aims at improving data retrieval and reuse via ontologies, Semantic Web technologies, and Linked Data for the ocean sciences. Cruises, in the sense of research expeditions, are central events for ocean scientists. Consequently, information about these cruises and the involved vessels has to be shared and made …

Enhancing Ocean Research Data Access, Cyndy Chandler, Robert Groman, Adam Shepherd, Molly Allison, Robert Arko, Yu Chen, Peter Fox, David Glover, Pascal Hitzler, Adam Leadbetter, Thomas Narock, Patrick West, Peter Wiebe

Computer Science and Engineering Faculty Publications

No abstract provided.

Interactive Visualization Of Grt And Biohts Data, Sara Gharabaghi, Thomas Wischgoll, Rhonda J. Vickery, Ross Smith, Leslie M. Blaha, Thomas Lamkin, Steven Kawamoto, Robert Trevino, Eric Bardes, Scott Tabar

Computer Science and Engineering Faculty Publications

The scope of this project is to provide better tools for statistical and informational visual analysis for High Throughput Screening of Biological Infectious Agents (BioHTS), General Recognition Theory (GRT) modeling, and areas where pipelines of unstructured datasets of all types must be analyzed. A parallel coordinates plot is one of the more effective visualization methods for visualizing multi variant data.

Semantic Entry Pairing For Improved Data Validation And Discovery, Adam Shepherd, Cyndy Chandler, Robert Arko, Yanning Chen, Adila Krisnadhi, Pascal Hitzler, Thomas Narock, Robert Groman, Shannon Rauch

Computer Science and Engineering Faculty Publications

No abstract provided.

What Information About Cardiovascular Diseases Do People Search Online?, Ashutosh Sopan Jadhav, Stephen Wu, Amit P. Sheth, Jyotishman Pathak

Kno.e.sis Publications

The objective of this study is to understand the types of health information (health topics) that users search online for Cardiovascular Diseases, by performing categorization of health search queries (from Mayoclinic.com) using UMLS MetaMap based on UMLS concepts and semantic types.

Description Logics, Adila Krisnadhi, Pascal Hitzler

Computer Science and Engineering Faculty Publications

Description logics (DLs) is a family of knowledge representation (KR) languages which represent knowledge in a domain of interest using formal, logic-based semantics through knowledge bases (KBs) containing general assertions of describing relevant concepts - hence, the term description - and specific assertions about individuals and relationships among them.

Ontology Design Patterns For Ocean Science Data Discovery, Pascal Hitzler

Computer Science and Engineering Faculty Publications

No abstract provided.

Location Prediction Of Twitter Users Using Wikipedia, Revathy Krishnamurthy, Pavan Kapanipathi, Amit P. Sheth, Krishnaprasad Thirunarayan

Kno.e.sis Publications

The mining of user generated content in social media has proven very effective in domains ranging from personalization and recommendation systems to crisis management. The knowledge of online users locations makes their tweets more informative and adds another dimension to their analysis. Existing approaches to predict the location of Twitter users are purely data-driven and require large training data sets of geo-tagged tweets. The collection and modelling process of tweets can be time intensive. To overcome this drawback, we propose a novel knowledge based approach that does not require any training data. Our approach uses information in Wikipedia, about cities …

Alignment And Dataset Identification Of Linked Data In Semantic Web, Kalpa Gunaratna, Sarasi Lalithsena, Amit P. Sheth

Kno.e.sis Publications

The Linked Open Data (LOD) cloud has gained significant attention in the Semantic Web community over the past few years. With rapid expansion in size and diversity, it consists of over 800 interlinked datasets with over 60 billion triples. These datasets encapsulate structured data and knowledge spanning over varied domains such as entertainment, life sciences, publications, geography, and government. Applications can take advantage of this by using the knowledge distributed over the interconnected datasets, which is not realistic to find in a single place elsewhere. However, two of the key obstacles in using the LOD cloud are the limited support …

Web Ontology Language (Owl), Kunal Sengupta, Pascal Hitzler

Computer Science and Engineering Faculty Publications

Web Ontology Language (OWL) is a core world wide web consortium [W3C] standard Knowledge representation language for the Semantic Web.

Why The Data Train Needs Semantic Rails, Krzysztof Janowicz, Frank Van Harmelen, James A. Hendler, Pascal Hitzler

Computer Science and Engineering Faculty Publications

While catchphrases such as big data, smart data, data intensive science, or smart dust highlight different aspects, they share a common theme: Namely, a shift towards a data-centric perspective in which the synthesis and analysis of data at an ever-increasing spatial, temporal, and thematic resolution promises new insights, while, at the same time, reducing the need for strong domain theories as starting points. In terms of the envisioned methodologies, those catchphrases tend to emphasize the role of predictive analytics, i.e., statistical techniques including data mining and machine learning, as well as supercomputing. Interestingly, however, while this perspective takes the availability …

Distributed Local Trust Propagation Model And Its Cloud-Based Implementation, Dharan Kumar Reddy Althuru

Browse all Theses and Dissertations

World Wide Web has grown rapidly in the last two decades with user generated content and interactions. Trust plays an important role in providing personalized content recommendations and in improving our confidence in various online interactions. We review trust propagation models in the context of social networks, semantic web, and recommender systems. With an objective to make trust propagation models more flexible, we propose several extensions to the trust propagation models that can be implemented as configurable parameters in the system. We implement Local Partial Order Trust (LPOT) model that considers trust as well as distrust ratings and perform evaluation …

Combating Integrity Attacks In Industrial Control Systems, Chad Arnold

Browse all Theses and Dissertations

Industrial Control Systems are vulnerable to integrity attacks because of connectivity to the external Internet and trusted internal networking components that can become compromised. Integrity attacks can be modeled, analyzed, and sometimes remedied by exploiting properties of physical devices and reasoning about the trust worthiness of ICS communication components.

Industrial control systems (ICS) monitor and control the processes of public utility that society depends on - the electric power grid, oil and gas pipelines, transportation, and water facilities. Attacks that impact the operations of these critical assets could have devastating consequences. The complexity and desire to interconnect ICS components have …

A Novel Synergistic Model Fusing Electroencephalography And Functional Magnetic Resonance Imaging For Modeling Brain Activities, Konstantinos Michalopoulos

Browse all Theses and Dissertations

Study of the human brain is an important and very active area of research. Unraveling the way the human brain works would allow us to better understand, predict and prevent brain related diseases that affect a significant part of the population. Studying the brain response to certain input stimuli can help us determine the involved brain areas and understand the mechanisms that characterize behavioral and psychological traits.

In this research work two methods used for the monitoring of brain activities, Electroencephalography (EEG) and functional Magnetic Resonance (fMRI) have been studied for their fusion, in an attempt to bridge together the …

Automated Complexity-Sensitive Image Fusion, Brian Patrick Jackson

Browse all Theses and Dissertations

To construct a complete representation of a scene with environmental obstacles such as fog, smoke, darkness, or textural homogeneity, multisensor video streams captured in diferent modalities are considered. A computational method for automatically fusing multimodal image streams into a highly informative and unified stream is proposed. The method consists of the following steps: 1. Image registration is performed to align video frames in the visible band over time, adapting to the nonplanarity of the scene by automatically subdividing the image domain into regions approximating planar patches

2. Wavelet coefficients are computed for each of the input frames in each modality …

The Properties Of Property Alignment On The Semantic Web, Michelle Andreen Cheatham

Browse all Theses and Dissertations

Ontology alignment is an important step in enabling computers to query and reason across the many linked datasets on the semantic web. This is a difficult challenge because the ontologies underlying different linked datasets can vary in terms of subject area coverage, level of abstraction, ontology modeling philosophy, and even language. The alignment approach presented here centers on string similarity metrics. Nearly all ontology alignment systems use a string similarity metric in one form or another, but it seems that the choice of a particular metric is often arbitrary. We begin this dissertation with the most comprehensive survey to date …

An Evolutionary Approximation To Contrastive Divergence In Convolutional Restricted Boltzmann Machines, Ryan R. Mccoppin

Browse all Theses and Dissertations

Deep learning is an emerging area in machine learning that exploits multi-layered neural networks to extract invariant relationships from large data sets. Deep learning uses layers of non-linear transformations to represent data in abstract and discrete forms. Several different architectures have been developed over the past few years specifically to process images including the Convolutional Restricted Boltzmann Machine. The Boltzmann Machine is trained using contrastive divergence, a depth-first gradient based training algorithm. Gradient based training methods have no guarantee of reaching an optimal solution and tend to search a limited region of the solution space. In this thesis, we present …

What Machines Understand About Personality Words After Reading The News, Eric David Moyer

Browse all Theses and Dissertations

Vector-based lexical semantics is a powerful technique that still has many undiscovered applications. In this thesis I apply a vector-space lexical-semantic model newly developed by Mikolov et. al. trained on skip-grams to the lexical hypothesis in personality psychology. The method produces interpretable dimensions that are consistent across several sets of descriptive personality words. The dimensions include ones for conflict and positive and negative evaluation. However they are more descriptive of word usage semantics than of the characteristics of the thing described and thus do not include a recognizable component of the 5 factor model in their first 14 dimensions. They …

Mining Privacy Settings To Find Optimal Privacy-Utility Tradeoffs For Social Network Services, Shumin Guo

Browse all Theses and Dissertations

Privacy has been a big concern for users of social network services (SNS). On recent criticism about privacy protection, most SNS now provide fine privacy controls, allowing users to set visibility levels for almost every profile item. However, this also creates a number of difficulties for users. First, SNS providers often set most items by default to the highest visibility to improve the utility of social network, which may conflict with users' intention. It is often formidable for a user to fine-tune tens of privacy settings towards the user desired settings. Second, tuning privacy settings involves an intricate tradeoff between …