Computer Sciences Commons^™

A Keyword Sense Disambiguation Based Approach For Noise Filtering In Twitter, Sanjaya Wijeratne, Bahareh R. Heravi

Kno.e.sis Publications

In this paper, we describe an approach to filter out noisy data generated by keywords-based tweet filtering methods by performing Word Sense Disambiguation on those keywords used to collect tweets. We present the noise filtering problem as a binary classification problem and discuss our evaluation strategy which is to be carried out in future.

Rasp-Qs: Efficient And Confidential Query Services In The Cloud, Zohreh S. Alavi, Lu Zhou, James L. Powers, Keke Chen

Kno.e.sis Publications

Hosting data query services in public clouds is an attractive solution for its great scalability and significant cost savings. However, data owners also have concerns on data privacy due to the lost control of the infrastructure. This demonstration shows a prototype for efficient and confidential range/kNN query services built on top of the random space perturbation (RASP) method. The RASP approach provides a privacy guarantee practical to the setting of cloudbased computing, while enabling much faster query processing compared to the encryption-based approach. This demonstration will allow users to more intuitively understand the technical merits of the RASP approach via …

Document Retrieval Using Predication Similarity, Kalpa Gunaratna

Kno.e.sis Publications

Document retrieval has been an important research problem over many years in the information retrieval community. State-of-the-art techniques utilize various methods in matching documents to a given document including keywords, phrases, and annotations. In this paper, we propose a new approach for document retrieval that utilizes predications (subject-predicate-object triples) extracted from the documents. We represent documents as sets of predications. We measure the similarity between predications to compute the similarity between documents. Our approach utilizes the hierarchical information available in ontologies in computing concept-concept similarity, making the approach flexible. Predication-based document similarity is more precise and forms the basis for …

A Novel Web-Based Depth Video Rewind Approach Toward Fall Preventive Interventions In Hospitals, Moein Enayati, Tanvi Banerjee, Mihail Popescu, Marjorie Skubic, Marilyn J. Rantz

Kno.e.sis Publications

Falls in the hospital rooms are considered a huge burden on healthcare costs. They can lead to injuries, extended length of stay, and increase in cost for both the patients and the hospital. It can also lead to emotional trauma for the patients and their families [1]. Having Microsoft Kinects installed in the hospital rooms to capture and process every movement in the room, we deployed our previously developed fall-detection system to detect naturally occurring falls, generate a real-time fall alarm and broadcast it to hospital nurses for immediate intervention. These systems also store a processed and reduced version …

Improving Structural Features Prediction In Protein Structure Modeling, Ashraf Yaseen

Computer Science Theses & Dissertations

Proteins play a vital role in the biological activities of all living species. In nature, a protein folds into a specific and energetically favorable three-dimensional structure which is critical to its biological function. Hence, there has been a great effort by researchers in both experimentally determining and computationally predicting the structures of proteins.

The current experimental methods of protein structure determination are complicated, time-consuming, and expensive. On the other hand, the sequencing of proteins is fast, simple, and relatively less expensive. Thus, the gap between the number of known sequences and the determined structures is growing, and is expected to …

Assisting Coordination During Crisis: A Domain Ontology Based Approach To Infer Resource Needs From Tweets, Shreyansh Bhatt, Hemant Purohit, Andrew J. Hampton, Valerie L. Shalin, Amit P. Sheth, John Flach

Kno.e.sis Publications

Ubiquitous social media during crises provides citizen reports on the situation, needs and supplies. Previous research extracts resource needs directly from the text (e.g. "Power cut to Coney Island and Brighton beach" indicates a power need). This approach assumes that citizens derive and write about specific needs from their observations, properly specified for the emergency response system, an assumption that is not consistent with general conversational behavior. In our study, Twitter messages (tweets) from Hurricane Sandy in 2012 clearly indicate power blackouts, but not their probable implications (e.g. loss of power to hospital life support systems). We use a domain …

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.

Risk Prediction With Genomic Data, Bharati Jadhav

Theses

Genome wide association study (GWAS) is widely used with various machine learning algorithms to predict disease risk. This thesis investigates this widely used approach of GWAS using Single Nucleotide Polymorphism (SNP) genotype data and a novel approach of disease risk prediction with whole exome sequencing data, namely Whole Exome Wide Association Study (WEWAS). It further applies a discriminating machine learning algorithm, namely a Support Vector Machine (SVM) with different Kernel functions. For this study, only SNPs generated using genotyping technology, which focuses more on common variants, are used initially for disease prediction. Later, the whole exome data generated using Next …

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 …

Modeling Stem Cell Population Dynamics, Samiur Arif

Computer Science Theses & Dissertations

Because of the stochastic nature of biological systems, mathematical and computational modeling approaches have become more acceptable to experimentalists and clinicians in recent years as contributing to new understandings of complicated cell mechanisms and tissue physiology. Indeed, even single cell or small tissue samples are complex dynamic systems that adapt to environmental challenges in space and time which is poorly understood. Mathematical models and computer simulations can explain and uncover unknown aspects of cell behavior and tissue functions. Models based on key biological mechanisms can give interesting insights and formulate predictions that cannot be derived from physical experiments or statistical …

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 …

Analysis Of Dna Motifs In The Human Genome, Yupu Liang

Dissertations, Theses, and Capstone Projects

DNA motifs include repeat elements, promoter elements and gene regulator elements, and play a critical role in the human genome. This thesis describes a genome-wide computational study on two groups of motifs: tandem repeats and core promoter elements.

Tandem repeats in DNA sequences are extremely relevant in biological phenomena and diagnostic tools. Computational programs that discover tandem repeats generate a huge volume of data, which can be difficult to decipher without further organization. A new method is presented here to organize and rank detected tandem repeats through clustering and classification. Our work presents multiple ways of expressing tandem repeats using …

Integrating Water Exclusion Theory Into Β Contacts To Predict Binding Free Energy Changes And Binding Hot Spots, Qian Liu, Steven C. H. Hoi, Chee Keong Kwoh, Limsoon Wong, Jinyan Li

Research Collection School Of Computing and Information Systems

Binding free energy and binding hot spots at protein-protein interfaces are two important research areas for understanding protein interactions. Computational methods have been developed previously for accurate prediction of binding free energy change upon mutation for interfacial residues. However, a large number of interrupted and unimportant atomic contacts are used in the training phase which caused accuracy loss. Results: This work proposes a new method, β ACV ASA , to predict the change of binding free energy after alanine mutations. β ACV ASA integrates accessible surface area (ASA) and our newly defined β contacts together into an atomic contact vector …

Integrating Water Exclusion Theory Into Β Contacts To Predict Binding Free Energy Changes And Binding Hot Spots, Qian Liu, Steven C. H. Hoi, Chee Keong Kwoh, Limsoon Wong, Jinyan Li

Research Collection School Of Computing and Information Systems

Binding free energy and binding hot spots at protein-protein interfaces are two important research areas for understanding protein interactions. Computational methods have been developed previously for accurate prediction of binding free energy change upon mutation for interfacial residues. However, a large number of interrupted and unimportant atomic contacts are used in the training phase which caused accuracy loss. Results: This work proposes a new method, β ACV ASA , to predict the change of binding free energy after alanine mutations. β ACV ASA integrates accessible surface area (ASA) and our newly defined β contacts together into an atomic contact vector …

Comparison Of Different Differential Expression Analysis Tools For Rna-Seq Data, Junfei Zhu

Theses

In molecular biology research, RNA-seq is a relatively new method for transcriptome profiling. It utilizes the next generation sequencing technology to provide huge amount information about the variety and abundance of RNA present in an organism of interest at a specific state and a given time. One of the most important tasks of RNA-seq analysis is finding genes that are expressed differently in different subject groups. A lot of differential expression analysis tools for RNA-seq have been developed, but there is no golden standard in this field. In this research, four commonly used tools (DESeq, edgeR, limma, and cuffdiff) are …

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.

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 …

Time Will Tell : Temporal Reasoning In Clinical Narratives And Beyond, Weiyi Sun

Legacy Theses & Dissertations (2009 - 2024)

Temporal reasoning in natural language refers to the extraction and understanding of time-related information conveyed in free text. A clinical narrative temporal reasoning component can enable a spectrum of medical natural language processing (NLP) applications that directly improve patient care documentation efficiency, accessibility and accountability. This dissertation contributes in three subtasks under temporal reasoning: temporal annotation, temporal expression extraction and temporal relation inferences. The temporal annotation work described in the dissertation produced one of the first publicly available clinical narratives. We published one of the first sets of temporal

The Rna Newton Polytope And Learnability Of Energy Parameters, Elmirasadat Forouzmand

Wayne State University Theses

Computational RNA secondary structure prediction has been a topic of much research interest for several decades now. Despite all the progress made in the field, even the state-of-the-art algorithms do not provide satisfying results, and the accuracy of output is limited for all the existent tools. Very complex energy models, different parameter estimation methods, and recent machine learning approaches had not been the answer for this problem. We believe that the first step to achieve results with high quality is to use the energy model with the potential for predicting accurate output. Hence, it is necessary to have a systematic …

De Novo Co-Assembly Of Bacterial Genomes From Multiple Single Cells, Narjes Sadat Movahedi Tabrizi

Wayne State University Theses

Recent progress in DNA amplication techniques, particularly multiple displacement amplication (MDA), has made it possible to sequence and assemble bacterial genomes from a single cell. However, the quality of single cell genome assembly has not yet reached the quality of normal multicell genome assembly due to the coverage bias and errors caused by MDA. Using a template of more than one cell for MDA or combining separate MDA products has been shown to improve the result of genome assembly from few single cells, but providing identical single cells, as a necessary step for these approaches, is a challenge. As a …

Identifying Aging-Related Genes In Mouse Hippocampus Using Gateway Nodes, Kathryn Dempsey Cooper, Hesham Ali

Interdisciplinary Informatics Faculty Publications

Background: High-throughput studies continue to produce volumes of metadata representing valuable sources of information to better guide biological research. With a stronger focus on data generation, analysis models that can readily identify actual signals have not received the same level of attention. This is due in part to high levels of noise and data heterogeneity, along with a lack of sophisticated algorithms for mining useful information. Networks have emerged as a powerful tool for modeling high-throughput data because they are capable of representing not only individual biological elements but also different types of relationships en masse. Moreover, well-established graph …

Teak: A Novel Computational And Gui Software Pipeline For Reconstructing Biological Networks, Detecting Activated Biological Subnetworks, And Querying Biological Networks., Thair Judeh

Wayne State University Dissertations

As high-throughput gene expression data becomes cheaper and cheaper, researchers are faced with a deluge of data from which biological insights need to be extracted and mined since the rate of data accumulation far exceeds the rate of data analysis. There is a need for computational frameworks to bridge the gap and assist researchers in their tasks. The Topology Enrichment Analysis frameworK (TEAK) is an open source GUI and software pipeline that seeks to be one of many tools that fills in this gap and consists of three major modules. The first module, the Gene Set Cultural Algorithm, de novo …

Algorithms And Tools For Computational Analysis Of Human Transcriptome Using Rna-Seq, Nan Deng

Wayne State University Dissertations

Alternative splicing plays a key role in regulating gene expression, and more than 90% of human genes are alternatively spliced through different types of alternative splicing. Dysregulated alternative splicing events have been linked to a number of human diseases. Recently, high-throughput RNA-Seq technologies have provided unprecedented opportunities to better characterize and understand transcriptomes, in particular useful for the detection of splicing variants between healthy and diseased human transcriptomes.

We have developed two novel algorithms and tools and a computational workflow to interrogate human transcriptomes between healthy and diseased conditions. The first is a read count-based Expectation-Maximization (EM) algorithm and tool, …