Computer Sciences Commons^™

Unsupervised Deep Structured Semantic Models For Commonsense Reasoning, Shuohang Wang, Sheng Zhang, Yelong Shen, Xiaodong Liu, Jingjing Liu, Jianfeng Gao, Jing Jiang

Research Collection School Of Computing and Information Systems

Commonsense reasoning is fundamental to natural language understanding. While traditional methods rely heavily on human-crafted features and knowledge bases, we explore learning commonsense knowledge from a large amount of raw text via unsupervised learning. We propose two neural network models based on the Deep Structured Semantic Models (DSSM) framework to tackle two classic commonsense reasoning tasks, Winograd Schema challenges (WSC) and Pronoun Disambiguation (PDP). Evaluation shows that the proposed models effectively capture contextual information in the sentence and co-reference information between pronouns and nouns, and achieve significant improvement over previous state-of-the-art approaches.

Enabling Multi-Hop Remote Method Invocation In Device-To-Device Networks, Minh Le, Stephen Clyde, Young‑Woo Kwon

Computer Science Faculty and Staff Publications

To avoid shrinking down the performance and preserve energy, low-end mobile devices can collaborate with the nearby ones by offloading computation intensive code. However, despite the long research history, code offloading is dilatory and unfit for applications that require rapidly consecutive requests per short period. Even though Remote Procedure Call (RPC) is apparently one possible approach that can address this problem, the RPC-based or message queue-based techniques are obsolete or unwieldy for mobile platforms. Moreover, the need of accessibility beyond the limit reach of the device-to-device (D2D) networks originates another problem. This article introduces a new software framework to overcome …

High-Performance Computing Frameworks For Large-Scale Genome Assembly, Sayan Goswami

LSU Doctoral Dissertations

Genome sequencing technology has witnessed tremendous progress in terms of throughput and cost per base pair, resulting in an explosion in the size of data. Typical de Bruijn graph-based assembly tools demand a lot of processing power and memory and cannot assemble big datasets unless running on a scaled-up server with terabytes of RAMs or scaled-out cluster with several dozens of nodes. In the first part of this work, we present a distributed next-generation sequence (NGS) assembler called Lazer, that achieves both scalability and memory efficiency by using partitioned de Bruijn graphs. By enhancing the memory-to-disk swapping and reducing the …

Using Big Data Analytics To Improve Hiv Medical Care Utilisation In South Carolina: A Study Protocol, Bankole Olatosi, Jiajia Zhang, Sharon Weissman, Jianjun Hu, Mohammad Rifat Haider, Xiaoming Li

Faculty Publications

Introduction Linkage and retention in HIV medical care remains problematic in the USA. Extensive health utilisation data collection through electronic health records (EHR) and claims data represent new opportunities for scientific discovery. Big data science (BDS) is a powerful tool for investigating HIV care utilisation patterns. The South Carolina (SC) office of Revenue and Fiscal Affairs (RFA) data warehouse captures individual-level longitudinal health utilisation data for persons living with HIV (PLWH). The data warehouse includes EHR, claims and data from private institutions, housing, prisons, mental health, Medicare, Medicaid, State Health Plan and the department of health and human services. The …

Context-Aware Wi-Fi Infrastructure-Based Indoor Positioning Systems, Huy Phuong Tran

Dissertations and Theses

Large enterprises are often interested in tracking objects and people within buildings to improve resource allocation and occupant experience. Infrastructure-based indoor positioning systems (IIPS) can provide this service at low-cost by leveraging already deployed Wi-Fi infrastructure. Typically, IIPS perform localization and tracking of devices by measuring only Wi-Fi signals at wireless access points and do not rely on inertial sensor data at mobile devices (e.g., smartphones), which would require explicit user consent and sensing capabilities of the devices.

Despite these advantages, building an economically viable cost-effective IIPS that can accurately and simultaneously track many devices over very large buildings is …

Field Drilling Data Cleaning And Preparation For Data Analytics Applications, Daniel Cardoso Braga

LSU Master's Theses

Throughout the history of oil well drilling, service providers have been continuously striving to improve performance and reduce total drilling costs to operating companies. Despite constant improvement in tools, products, and processes, data science has not played a large part in oil well drilling. With the implementation of data science in the energy sector, companies have come to see significant value in efficiently processing the massive amounts of data produced by the multitude of internet of thing (IOT) sensors at the rig. The scope of this project is to combine academia and industry experience to analyze data from 13 different …

Weaving The Dark Web: Legitimacy On Freenet, Tor, And I2p, John Schriner

Publications and Research

This is a book review of Robert W. Gehl's Weaving the Dark Web: Legitimacy on Freenet, Tor, and I2P (2018). The book explores these anonymity networks and the concept of legitimacy throughout. Using a multidisciplinary approach and interviews with network-builders and users, Gehl helps to demystify the dark web and critically examine these networks and technologies.

Implementation Of Multivariate Artificial Neural Networks Coupled With Genetic Algorithms For The Multi-Objective Property Prediction And Optimization Of Emulsion Polymers, David Chisholm

Master's Theses

Machine learning has been gaining popularity over the past few decades as computers have become more advanced. On a fundamental level, machine learning consists of the use of computerized statistical methods to analyze data and discover trends that may not have been obvious or otherwise observable previously. These trends can then be used to make predictions on new data and explore entirely new design spaces. Methods vary from simple linear regression to highly complex neural networks, but the end goal is similar. The application of these methods to material property prediction and new material discovery has been of high interest …

Decentralise Me, Paul Robert Griffin

MITB Thought Leadership Series

If you are in need of a reminder of the levels of hype surrounding blockchain, then look no further than Japan’s J-Pop sensation Kasotsuka Shojo, otherwise known as the Virtual Currency Girls who, with their debut track “The Moon and Cryptocurrencies and Me”, aim to educate fans about cryptocurrencies in an entertaining way.

In The Absence Of Information, 1/N Investment Makes Perfect Sense, Julio Urenda, Vladik Kreinovich

Departmental Technical Reports (CS)

When people have several possible investment instruments, people often invest equally into these instruments: in the case of n instruments, they invest 1/n of their money into each of these instruments. Of course, if additional information about each instrument is available, this 1/n investment strategy is not optimal. We show, however, that in the absence of reliable information, 1/n investment is indeed the best strategy.

Faster Quantum Alternative To Softmax Selection In Deep Learning And Deep Reinforcement Learning, Oscar Galindo, Christian Ayub, Martine Ceberio, Vladik Kreinovich

Departmental Technical Reports (CS)

Deep learning and deep reinforcement learning are, at present, the best available machine learning tools for use in engineering problems. However, at present, the use of these tools is limited by the fact that they are very time-consuming, usually requiring the use of a high performance computer. It is therefore desirable to look for possible ways to speed up the corresponding computations. One of the time-consuming parts of these algorithms is softmax selection, when instead of selecting the alternative with the largest possible value of the corresponding objective function, we select all possible values, with probabilities increasing with the value …

Closing Banquet Eulogies, Russell Howell, C. Ray Rosentrater

ACMS Conference Proceedings 2019

A tribute to David Lay; A tribute to John Roe

Evaluating The Efficacy Of Magnetometer-Based Vehicle Sensors, Luke A. Hudspeth

Dartmouth College Undergraduate Theses

The issue of parking is more at the forefront of urban development than one might believe. In fact, academic studies have shown that roughly 30% of city traffic is due to drivers circling city blocks attempting to find an open spot. Due to such congestion people often avoid urban centers and downtown areas for shopping or dining because parking is such a hassle and assumed to be unavailable. If drivers knew where parking was available in real time, they could proceed directly to open spaces as opposed to their congestion-inducing attempts to park. A better solution would guide drivers to …

Bincor: An R Package For Estimating The Correlation Between Two Unevenly Spaced Time Series, Josue M. Polanco-Martinez, Martin A. Medina-Elizalde, Maria Fernanda Sanchez Goni, Manfred Mudelsee

The R Journal

This paper presents a computational program named BINCOR (BINned CORrelation) for estimating the correlation between two unevenly spaced time series. This program is also applicable to the situation of two evenly spaced time series not on the same time grid. BINCOR is based on a novel estimation approach proposed by Mudelsee (2010) for estimating the correlation between two climate time series with different timescales. The idea is that autocorrelation (e.g. an AR1 process) means that memory enables values obtained on different time points to be correlated. Binned correlation is performed by resampling the time series under study into time bins …

”Cyberworld” As A Theme For A University-Wide First-Year Common Course, Kristen Przyborski, Frank Breitinger, Lauren Beck, Ronald S. Harichandran

Engineering and Applied Science Education Faculty Publications

Nowadays we all live in a cyber world and use the internet for emailing, banking, streaming video, shopping, reading news, or other activities. Given all the time people spend online, it is important that all students (regardless of their major) learn some basics about living in a cyber world, e.g., strategies for online safety, impact of artificial intelligence, digital forensics or ancestry.com. To facilitate students from many majors to learn about important issues related to the internet, eight faculty from a variety of disciplines at the University of New Haven integrated the theme of Cyber World into our team-taught, first-year …

Forecasting Building Energy Consumption With Deep Learning: A Sequence To Sequence Approach, Ljubisa Sehovac, Cornelius Nesen, Katarina Grolinger

Electrical and Computer Engineering Publications

Energy Consumption has been continuously increasing due to the rapid expansion of high-density cities, and growth in the industrial and commercial sectors. To reduce the negative impact on the environment and improve sustainability, it is crucial to efficiently manage energy consumption. Internet of Things (IoT) devices, including widely used smart meters, have created possibilities for energy monitoring as well as for sensor based energy forecasting. Machine learning algorithms commonly used for energy forecasting such as feedforward neural networks are not well-suited for interpreting the time dimensionality of a signal. Consequently, this paper uses Recurrent Neural Networks (RNN) to capture time …

How To Use Quantum Computing To Check Which Inputs Are Relevant: A Proof That Deutsch-Jozsa Algorithm Is, In Effect, The Only Possibility, Vladik Kreinovich, Martine Ceberio, Ricardo Alvarez

Departmental Technical Reports (CS)

One of the main reasons why computations -- in particular, engineering computations -- take long is that, to be on the safe side, models take into account all possible affecting features, most of which turn out to be not really relevant for the corresponding physical problem. From this viewpoint, it is desirable to find out which inputs are relevant. In general, the problem of checking the input's relevancy is itself NP-hard, which means, crudely speaking, that no feasible algorithm can always solve it. Thus, it is desirable to speed up this checking as much as possible. One possible way to …

Intuitive Idea Of Implication Vs. Formal Definition: How To Define The Corresponding Degree, Olga Kosheleva, Vladik Kreinovich

Departmental Technical Reports (CS)

Formal implication does not capture the intuitive idea of "if A then B", since in formal implication, every two true statements -- even completely unrelated ones -- imply each other. A more adequate description of intuitive implication happens if we consider how much the use of A can shorten a derivation of B. At first glance, it may seem that the number of bits by which we shorten this derivation is a reasonable degree of implication, but we show that this number is not in good accordance with our intuition, and that a natural formalization of this intuition leads to …

Exploring The Dynamics Of Scientific Research, Shilpa Lakhanpal

Dissertations

Scientific research papers present the research endeavors of numerous scientists around the world, and are documented across multitudes of technical conference proceedings, and other such publications. Given the plethora of such research data, if we could automate the extraction of key interesting areas of research, and provide access to this new information, it would make literature searches incredibly easier for researchers. This in turn could be very useful for them in furthering their research agenda. With this goal in mind, we have endeavored to provide such solutions through our research. Specifically, the focus of our research is to design, analyze …

A Study On The Mental Health Of Women In The Csse Department Of California Polytechnic State University, Sophia Lao, Gabriella Garcia Nobili

Computer Science and Software Engineering

The goal of this paper is to explore the perceived effect that the Computer Science Curriculum has on the mental health of female students. To discover these effects we conducted 50 interviews over a period of several months with the female students of the Computer Science and Software Engineering (CSSE) department at California Polytechnic State University in San Luis Obispo. The total amount of Computer Science (CSC) majors at this institution is approximately 680, with 150 of them being female, so we were able to gather responses from 33.3% of the female CSC students from our interviews. It is worth …

Enhancing The Study Of Art History Utilizing Computational Thinking: Focus, Abstraction In Art (Modern), Douglas L. Moody, Sandra Cheng

Open Educational Resources

No abstract provided.

Changes On Cran, Kurt Hornik, Uwe Ligges, Achim Zeileis

The R Journal

In the past 8 months, 1524 new packages were added to the CRAN package repository. 71 packages were unarchived and 302 were archived. The following shows the growth of the number of active packages in the CRAN package repository:

R Foundation News, Torsten Hothorn

The R Journal

Membership fees and donations received between 2019-01-07 and 2019-09-04

R Package For Geometric Shadow Calculations In An Urban Environment, Michael Dorman, Evyatar Erell, Adi Vulkan, Itai Kloog

The R Journal

This paper introduces the shadow package for R. The package provides functions for shadow-related calculations in the urban environment, namely shadow height, shadow footprint and Sky View Factor (SVF) calculations, as well as a wrapper function to estimate solar radiation while taking shadow effects into account. All functions operate on a layer of polygons with a height attribute, also known as “extruded polygons” or 2.5D vector data. Such data are associated with accuracy limitations in representing urban environments. However, unlike 3D models, polygonal layers of building outlines along with their height are abundantly available and their processing does not require …

Indoor Positioning And Fingerprinting: The R Package Ipft, Emilio Sansano, Raúl Montoliu, Óscar Belmonte, Joaquín Torres-Sospedra

The R Journal

Methods based on Received Signal Strength Indicator (RSSI) fingerprinting are in the forefront among several techniques being proposed for indoor positioning. This paper introduces the R package ipft, which provides algorithms and utility functions for indoor positioning using fingerprinting techniques. These functions are designed for manipulation of RSSI fingerprint data sets, estimation of positions, comparison of the performance of different positioning models, and graphical visualization of data. Well-known machine learning algorithms are implemented in this package to perform analysis and estimations over RSSI data sets. The paper provides a description of these algorithms and functions, as well as examples of …

Time-Series Clustering In R Using The Dtwclust Package, Alexis Sardá-Espinosa

The R Journal

Most clustering strategies have not changed considerably since their initial definition. The common improvements are either related to the distance measure used to assess dissimilarity, or the function used to calculate prototypes. Time-series clustering is no exception, with the Dynamic Time Warping distance being particularly popular in that context. This distance is computationally expensive, so many related optimizations have been developed over the years. Since no single clustering algorithm can be said to perform best on all datasets, different strategies must be tested and compared, so a common infrastructure can be advantageous. In this manuscript, a general overview of shape-based …

Editorial, Norm Matloff

The R Journal

On behalf of the Editorial Board, I am pleased to present Volume 10, Issue 1 of the R Journal. This issue contains 36 contributed articles. The majority of which cover new or newly enhanced packages on CRAN.

Optimization Routines For Enforcing One-To-One Matches In Record Linkage Problems, Diego Moretti, Luca Valentino, Tiziana Tuoto

The R Journal

Record linkage aims at quickly and accurately identifying if two records represent the same real world entity. In many applications, we are interested in restricting the linkage results to "1 to 1" links, that is a single record does not appear more than once in the output. This can be dealt with the transport algorithm. The optimization problem, however, grows quadratically in the size of the input, quickly becoming untreatable for cases with a few thousand records. This paper compares different solutions, provided by some R packages for linear programming solvers. The comparison is done in terms of memory usage …

Fclust: An R Package For Fuzzy Clustering, Maria Brigida Ferraro, Paolo Giordani, Alessio Serafini

The R Journal

Fuzzy clustering methods discover fuzzy partitions where observations can be softly assigned to more than one cluster. The package fclust is a toolbox for fuzzy clustering in the R programming language. It not only implements the widely used fuzzy k-means (FkM) algorithm, but also many FkM variants. Fuzzy cluster similarity measures, cluster validity indices and cluster visualization tools are also offered. In the current version, all the functions are rewritten in the C++ language allowing their application in large-size problems. Moreover, new fuzzy relational clustering algorithms for partitioning qualitative/mixed data are provided together with an improved version of the so-called …

Mdfs: Multidimensional Feature Selection In R, Radosław Piliszek, Krzysztof Mnich, Szymon Migacz, Paweł Tabaszewski, Andrzej Sułecki, Aneta Polewko-Klim, Witold Rudnicki

The R Journal

Identification of informative variables in an information system is often performed using simple one-dimensional filtering procedures that discard information about interactions between variables. Such an approach may result in removing some relevant variables from consideration. Here we present an R package MDFS (MultiDimensional Feature Selection) that performs identification of informative variables taking into account synergistic interactions between multiple descriptors and the decision variable. MDFS is an implementation of an algorithm based on information theory (Mnich and Rudnicki, 2017). The computational kernel of the package is implemented in C++. A high-performance version implemented in CUDA C is also available. The application …