Computer Sciences Commons^™

Enriching Language Models With Visually-Grounded Word Vectors And The Lancaster Sensorimotor Norms, Casey Kennington

Computer Science Faculty Publications and Presentations

Language models are trained only on text despite the fact that humans learn their first language in a highly interactive and multimodal environment where the first set of learned words are largely concrete, denoting physical entities and embodied states. To enrich language models with some of this missing experience, we leverage two sources of information: (1) the Lancaster Sensorimotor norms, which provide ratings (means and standard deviations) for over 40,000 English words along several dimensions of embodiment, and which capture the extent to which something is experienced across 11 different sensory modalities, and (2) vectors from coefficients of binary classifiers …

An Analysis On Pixel Redundancy Structure In Equirectangular Images, I. Vazquez, S. Cutchin

Computer Science Faculty Publications and Presentations

360° photogrammetry captures the surrounding light from a central point. To process and transmit these types of images over the network to the end user, the most common approach is to project them onto a 2D image using the equirectangular projection to generate a 360° image. However, this projection introduces redundancy into the image, increasing storage and transmission requirements. To address this problem, the standard approach is to use compression algorithms, such as JPEG or PNG, but they do not take full advantage of the visual redundancy produced by the equirectangular projection. In this study of the 360SP dataset (a …

Texture Classification Using Angular And Radial Bins In Transformed Domain, Arun D. Kulkarni, Aavash Sthapit, Ashim Sedhain, Bishrut Bhattarai, Saurav Panthee

Computer Science Faculty Publications and Presentations

Texture is generally recognized as fundamental to perceptions. There is no precise definition or characterization available in practice. Texture recognition has many applications in areas such as medical image analysis, remote sensing, and robotic vision. Various approaches such as statistical, structural, and spectral have been suggested in the literature. In this paper we propose a method for texture feature extraction. We transform the image into a two-dimensional Discrete Cosine Transform (DCT) and extract features using the ring and wedge bins in the DCT plane. These features are based on texture properties such as coarseness, smoothness, graininess, and directivity of the …

Enhancing Classroom Instruction With Online News, Michael D. Ekstrand, Katherine Landau Wright, Maria Soledad Pera

Computer Science Faculty Publications and Presentations

Purpose

Investigate how school teachers look for informational texts for their classrooms. Access to current, varied, and authentic informational texts improves learning outcomes for K-12 students, but many teachers lack resources to expand and update readings. The Web offers freely-available resources, but finding suitable ones is time-consuming. This research lays the groundwork for building tools to ease that burden.

Methodology

This paper reports qualitative findings from a study in two stages: (1) a set of semi-structured interviews, based on the Critical Incident Technique, eliciting teachers’ information-seeking practices and challenges; and (2) observations of teachers using a prototype teaching-oriented news search …

Lenskit For Python: Next-Generation Software For Recommender Systems Experiments, Michael D. Ekstrand

Computer Science Faculty Publications and Presentations

LensKit is an open-source toolkit for building, researching, and learning about recommender systems. First released in 2010 as a Java framework, it has supported diverse published research, small-scale production deployments, and education in both MOOC and traditional classroom settings. In this paper, I present the next generation of the LensKit project, re-envisioning the original tool's objectives as flexible Python package for supporting recommender systems research and development. LensKit for Python (LKPY) enables researchers and students to build robust, flexible, and reproducible experiments that make use of the large and growing PyData and Scientific Python ecosystem, including scikit-learn, and TensorFlow. To …

Evaluating Stochastic Rankings With Expected Exposure, Fernando Diaz, Bhaskar Mitra, Michael D. Ekstrand, Asia J. Biega, Ben Carterette

Computer Science Faculty Publications and Presentations

We introduce the concept of expected exposure as the average attention ranked items receive from users over repeated samples of the same query. Furthermore, we advocate for the adoption of the principle of equal expected exposure: given a fixed information need, no item should receive more or less expected exposure than any other item of the same relevance grade. We argue that this principle is desirable for many retrieval objectives and scenarios, including topical diversity and fair ranking. Leveraging user models from existing retrieval metrics, we propose a general evaluation methodology based on expected exposure and draw connections to related …

Falcon: Framework For Anomaly Detection In Industrial Control Systems, Subin Sapkota, A.K.M. Nuhil Mehdy, Stephen Reese, Hoda Mehrpouyan

Computer Science Faculty Publications and Presentations

Industrial Control Systems (ICS) are used to control physical processes in critical infrastructure. These systems are used in a wide variety of operations such as water treatment, power generation and distribution, and manufacturing. While the safety and security of these systems are of serious concern, recent reports have shown an increase in targeted attacks aimed at manipulating physical processes to cause catastrophic consequences. This trend emphasizes the need for algorithms and tools that provide resilient and smart attack detection mechanisms to protect ICS. In this paper, we propose an anomaly detection framework for ICS based on a deep neural network. …

Ieee Access Special Section Editorial: Machine Learning Designs, Implementations And Techniques, Shadi A. Aljawarneh, Oguz Bayat, Juan A. Lara, Robert P. Schumaker

Computer Science Faculty Publications and Presentations

IEEE access special section editorial.

Rrsds: Towards A Robot-Ready Spoken Dialogue System, Casey Kennington, Daniele Moro, Lucas Marchand, Jake Carns, David Mcneill

Computer Science Faculty Publications and Presentations

Spoken interaction with a physical robot requires a dialogue system that is modular, multimodal, distributive, incremental and temporally aligned. In this demo paper, we make significant contributions towards fulfilling these requirements by expanding upon the ReTiCo incremental framework. We outline the incremental and multimodal modules and how their computation can be distributed. We demonstrate the power and flexibility of our robot-ready spoken dialogue system to be integrated with almost any robot.

Semantics With Feeling: Emotions For Abstract Embedding, Affect For Concrete Grounding, Daniele Moro, Gerardo Caracas, David Mcneill, Casey Kennington

Computer Science Faculty Publications and Presentations

An important yet underexplored aspect of meaning in both distributional and grounded models of semantics is emotion. In this paper, we explore how emotion can be predicted from descriptions of robot behaviors represented with embeddings. We then compare this approach with a grounded model that maps corresponding robot behaviors represented as internal states to the same emotion labels and discover comparable results. We then take the predictions from the second model and use them as a proxy for concrete affect (as opposed to abstract emotion) and use this derived affect to ground a semantic classifier in a retrieval task and …

Learning Word Groundings From Humans Facilitated By Robot Emotional Displays, David Mcneill, Casey Kennington

Computer Science Faculty Publications and Presentations

In working towards accomplishing a human-level acquisition and understanding of language, a robot must meet two requirements: the ability to learn words from interactions with its physical environment, and the ability to learn language from people in settings for language use, such as spoken dialogue. In a live interactive study, we test the hypothesis that emotional displays are a viable solution to the cold-start problem of how to communicate without relying on language the robot does not–indeed, cannot–yet know. We explain our modular system that can autonomously learn word groundings through interaction and show through a user study with 21 …

Estimating Error And Bias In Offline Evaluation Results, Mucun Tian, Michael D. Ekstrand

Computer Science Faculty Publications and Presentations

Offline evaluations of recommender systems attempt to estimate users’ satisfaction with recommendations using static data from prior user interactions. These evaluations provide researchers and developers with first approximations of the likely performance of a new system and help weed out bad ideas before presenting them to users. However, offline evaluation cannot accurately assess novel, relevant recommendations, because the most novel items were previously unknown to the user, so they are missing from the historical data and cannot be judged as relevant.

We present a simulation study to estimate the error that such missing data causes in commonly-used evaluation metrics in …

Developing Big Data Projects In Open University Engineering Courses: Lessons Learned, Juan A. Lara, Aurea Anguera De Sojo, Shadi Aljawarneh, Robert P. Schumaker, Bassam Al-Shargabi

Computer Science Faculty Publications and Presentations

Big Data courses in which students are asked to carry out Big Data projects are becoming more frequent as a part of University Engineering curriculum. In these courses, instructors and students must face a series of special characteristics, difficulties and challenges that it is important to know about beforehand, so the lecturer can better plan the subject and manage the teaching methods in order to prevent students' academic dropout and low performance. The goal of this research is to approach this problem by sharing the lessons learned in the process of teaching e-learning courses where students are required to develop …

Defect-Free Plastic Deformation Through Dimensionality Reduction And Self-Annihilation Of Topological Defects In Crystalline Solids, Yipeng Gao, Yongfeng Zhang, Larry K. Aagesen, Jianguo Yu, Min Long, Yunzhi Wang

Computer Science Faculty Publications and Presentations

As a signature of symmetry-breaking processes, the generation and annihilation of topological defects (domain walls, strings, etc.) are of great interest in condensed matter physics and cosmology. Here we propose a distinctive self-organization process through phase transitions, in which all the generated topological defects are dimensionality reduced and self-annihilated. In crystalline solids, such a unique mechanism allows a perfect single crystal after plastic deformation, which originates from the coupling of different types of broken symmetries.

We’Ve Only Just Begun: Children Searching In The Classroom, Monica Landoni, Theo Huibers, Emiliana Murgia, Maria Soledad Pera

Computer Science Faculty Publications and Presentations

In this extended abstract, we present an overview of our ongoing project. Specifically, we briefly discuss the motivation for our research agenda, research goals in the short and long term, and the body of work we have published thus far that serves as the foundation upon which we build the next steps related to Information Retrieval and Children in the Classroom Setting.

Kidspell: A Child-Oriented, Rule-Based, Phonetic Spellchecker, Brody Downs, Oghenemaro Anuyah, Aprajita Shukla, Jerry Alan Fails, Maria Soledad Pera, Katherine Wright, Casey Kennington

Computer Science Faculty Publications and Presentations

For help with their spelling errors, children often turn to spellcheckers integrated in software applications like word processors and search engines. However, existing spellcheckers are usually tuned to the needs of traditional users (i.e., adults) and generally prove unsatisfactory for children. Motivated by this issue, we introduce KidSpell, an English spellchecker oriented to the spelling needs of children. KidSpell applies (i) an encoding strategy for mapping both misspelled words and spelling suggestions to their phonetic keys and (ii) a selection process that prioritizes candidate spelling suggestions that closely align with the misspelled word based on their respective keys. To assess …

On Basing One-Way Permutations On Np-Hard Problems Under Quantum Reductions, Nai-Hui Chia, Sean Hallgren, Fang Song

Computer Science Faculty Publications and Presentations

A fundamental pursuit in complexity theory concerns reducing worst-case problems to average-case problems. There exist complexity classes such as PSPACE that admit worst-case to average-case reductions. However, for many other classes such as NP, the evidence so far is typically negative, in the sense that the existence of such reductions would cause collapses of the polynomial hierarchy(PH). Basing cryptographic primitives, e.g., the average-case hardness of inverting one-way permutations, on NP-completeness is a particularly intriguing instance. As there is evidence showing that classical reductions from NP-hard problems to breaking these primitives result in PH collapses, it seems unlikely to base cryptographic …

Selectivity And Robustness Of Sparse Coding Networks, Dylan M. Paiton, Charles Frye, Sheng Y. Lundquist, Joel D. Bowen, Ryan Zarcone, Bruno A. Olshausen

Computer Science Faculty Publications and Presentations

We investigate how the population nonlinearities resulting from lateral inhibition and thresholding in sparse coding networks influence neural response selectivity and robustness. We show that when compared to pointwise nonlinear models, such population nonlinearities improve the selectivity to a preferred stimulus and protect against adversarial perturbations of the input. These findings are predicted from the geometry of the single-neuron iso-response surface, which provides new insight into the relationship between selectivity and adversarial robustness. Inhibitory lateral connections curve the iso-response surface outward in the direction of selectivity. Since adversarial perturbations are orthogonal to the iso-response surface, adversarial attacks tend to be …

Computer Science For Equity: Teacher Education, Agency, And Statewide Reform, Joanna Goode, Max Skorodinsky, Jill Hubbard, James Hook

Computer Science Faculty Publications and Presentations

This paper reports on a statewide “Computer Science for All” initiative in Oregon that aims to democratize high school computer science and broaden participation in an academic subject that is one of the most segregated disciplines nationwide, in terms of both race and gender. With no statewide policies to support computing instruction, Oregon's legacy of computer science education has been marked by both low participation and by rates of underrepresented students falling well-below the already dismal national rates. The study outlined in this paper focuses on how teacher education can support educators in developing knowledge and agency, and impacting policies …

Evaluating And Improving Child-Directed Automatic Speech Recognition, Eric Booth, Jake Carns, Casey Kennington, Nader Rafla

Computer Science Faculty Publications and Presentations

Speech recognition has seen dramatic improvements in the last decade, though those improvements have focused primarily on adult speech. In this paper, we assess child-directed speech recognition and leverage a transfer learning approach to improve child-directed speech recognition by training the recent DeepSpeech2 model on adult data, then apply additional tuning to varied amounts of child speech data. We evaluate our model using the CMU Kids dataset as well as our own recordings of child-directed prompts. The results from our experiment show that even a small amount of child audio data improves significantly over a baseline of adult-only or child-only …

A User-Centric And Sentiment Aware Privacy-Disclosure Detection Framework Based On Multi-Input Neural Network, A. K. M. Nuhil Mehdy, Hoda Mehrpouyan

Computer Science Faculty Publications and Presentations

Data and information privacy is a major concern of today’s world. More specifically, users’ digital privacy has become one of the most important issues to deal with, as advancements are being made in information sharing technology. An increasing number of users are sharing information through text messages, emails, and social media without proper awareness of privacy threats and their consequences. One approach to prevent the disclosure of private information is to identify them in a conversation and warn the dispatcher before the conveyance happens between the sender and the receiver. Another way of preventing information (sensitive) loss might be to …

Detecting Fake News Spreaders In Social Networks Via Linguistic And Personality Features: Notebook For Pan At Clef 2020, Anu Shrestha, Francesca Spezzano, Abishai Joy

Computer Science Faculty Publications and Presentations

This paper addresses the problem of automatically detecting fake news spreaders in social networks such as Twitter. We model the problem as a binary classification task and consider several groups of features, including writing style, word and char n-grams, BERT semantic embedding, and sentiment analysis, which are computed from a set of tweets each user authored. Our proposed approach is evaluated on the dataset made available by the PAN at CLEF 2020 shared task on profiling fake news spreader, which provided labeled data in both English and Spanish. Experimental results show that we can detect fake news spreaders with an …

Recurrent Neural Network Properties And Their Verification With Monte Carlo Techniques, Dmitry Vengertsev, Elena Sherman

Computer Science Faculty Publications and Presentations

As RNNs find its applications in medical and automotive fields, they became a part of critical systems, which traditionally require thorough verification processes. In this work we present how RNNs behaviors can be modeled as labeled transition systems and formally define a set of state and temporal safety properties for such models. To verify those properties we propose to use the Monte Carlo approach and evaluate its effectiveness for different type of properties. We perform empirical evaluation on two RNN models to determine to what extent they satisfy the properties and how many the samples of Monte Carlo required to …

What Snippets Feel: Depression, Search, And Snippets, Ashlee Milton, Maria Soledad Pera

Computer Science Faculty Publications and Presentations

Mental health disorders (MHD) is a rising, yet stigmatized, topic in the United States. Individuals suffering from MHD are slowing starting to overcome this stigma by discussing how technology affects them. Researchers have explored behavioral nuances that emerge from interactions of individuals affected by MHD with persuasive technologies, mainly social media. Yet, there is a gap in the analysis pertaining to search engines, another persuasive technology, which is part of their everyday lives. In this paper, we report the results of an initial exploratory analysis conducted to understand the sentiment/emotion profiles of search engines handling the information needs of searchers …

Say It With Emojis: Co-Designing Relevance Cues For Searching In The Classroom, Mohammad Aliannejadi, Monica Landoni, Theo Huibers, Emiliana Murgia, Maria Soledad Pera

Computer Science Faculty Publications and Presentations

Search Engine Result Pages (SERP) include snippets of retrieved resources as a means to help searchers select the ones that satisfy their information needs. This way, result relevance can be determined by scanning through snippets, an exercise that requires experience with reading, understanding, and assessing the value of a document. These are skills that primary school children are still developing and thus are not yet proficient with. As web search tools are essential to support children learning at school and home, we explore how to help young searchers in making informed relevance assessments while conducting searches in a classroom. In …

Prediction Of Fatality Crashes With Multilayer Perceptron Of Crash Record Information System Datasets, Thanh Hung Duong, Fengxiang Qiao, Jyh-Haw Yeh, Yunpeng Zhang

Computer Science Faculty Publications and Presentations

Despite the effort of the authorities and researchers, there has been no sign of decreasing in the number of fatal crashes annually. To analyze the deadly collisions, researchers have focused on finding which factors affect injury severity, and thus many crash prediction models for it had been developed. Commonly the injury severity is categorized into five different classes. Still, in many studies, minority classes like fatality and incapacitating injury were merged so that the dataset becomes balanced, and the model can provide decent predictions. However, this approach does not help analyze the fatal crashes as they are joined with other …

The Computer Science Professional's Hatchery, Amit Jain, Noah Salzman, Donald Winiecki

Computer Science Faculty Publications and Presentations

As a recipient of a National Science Foundation Revolutionizing Engineering and Computer Science Departments (RED) grant, the Computer Science Department at the Boise State University is building a Computer Science (CS) Professionals Hatchery. This paper is a summary to accompany the poster to be presented.

Event Trend Aggregation Under Rich Event Matching Semantics, Olga Poppe, Chuan Lei, Elke A. Rundensteiner, David Maier

Computer Science Faculty Publications and Presentations

Streaming applications from cluster monitoring to algorithmic trading deploy Kleene queries to detect and aggregate event trends. Rich event matching semantics determine how to compose events into trends. The expressive power of stateof- the-art streaming systems remains limited since they do not support many of these semantics. Worse yet, they suffer from long delays and high memory costs because they maintain aggregates at a fine granularity. To overcome these limitations, our Coarse-Grained Event Trend Aggregation (Cogra) approach supports a rich variety of event matching semantics within one system. Better yet, Cogra incrementally maintains aggregates at the coarsest granularity possible for …

Cross-Validating Traffic Speed Measurements From Probe And Stationary Sensors Through State Reconstruction, Jia Li, Kenneth Perrine, Lidong Wu, C. Michael Walton

Computer Science Faculty Publications and Presentations

Traffic speed on freeways can be measured by two types of technologies, i.e. probe sensors and stationary sensors. Cross-validation is critical to ensure the consistency between heterogeneous measurements. A challenge lies in the mismatch of probe and stationary measurements in space and time, especially when one of them is relatively sparse. Towards filling the gap, this paper presents a cross-validation method based on traffic state reconstruction. The proposed method is computationally simple and robust. This makes it ready to be implemented for large data sets without complicated tuning. We present analytical formulation of the proposed method and an analysis of …

Good Similar Patches For Image Denoising, Si Lu

Computer Science Faculty Publications and Presentations

Patch-based denoising algorithms like BM3D have achieved outstanding performance. An important idea for the success of these methods is to exploit the recurrence of similar patches in an input image to estimate the underlying image structures. However, in these algorithms, the similar patches used for denoising are obtained via Nearest Neighbour Search (NNS) and are sometimes not optimal. First, due to the existence of noise, NNS can select similar patches with similar noise patterns to the reference patch. Second, the unreliable noisy pixels in digital images can bring a bias to the patch searching process and result in a loss …