Search

This paper is interested in investigating whether human gaze signals can be leveraged to improve state-of-the-art search engine performance and how to incorporate this new input signal marked by human attention into existing neural retrieval models. In this paper, we propose GazBy ({\bf Gaz}e-based {\bf B}ert model for document relevanc{\bf y}), a light-weight joint model that integrates human gaze fixation estimation into transformer models to predict document relevance, incorporating more nuanced information about cognitive processing into information retrieval (IR). We evaluate our model on the Text Retrieval Conference (TREC) Deep Learning (DL) 2019 and 2020 Tracks. Our experiments show encouraging results and illustrate the effective and ineffective entry points for using human gaze to help with transformer-based neural retrievers. With the rise of virtual reality (VR) and augmented reality (AR), human gaze data will become more available. We hope this work serves as a first step exploring using gaze signals in modern neural search engines.

Data driven approaches to problem solving are, in many regards, the holy grail of evidence backed decision making. Using first-party empirical data to analyze behavior and establish predictions yields us the ability to base in-depth analyses on particular individuals and reduce our dependence on generalizations. Modern mobile and embedded devices provide a wealth of sensors and means for collecting and tracking individualized data. Applying these assets to the realm of insurance (which is a statistically backed endeavor at heart) is certainly nothing new; yet doing so in a way that is privacy-driven and secure has not been a central focus of implementers. Existing data-driven insurance technologies require a certain level of trust in the data tracking agency (i.e. insurer) to not misuse, mishandle, or over-collect user data. Smart contracts and blockchain technology provide us an opportunity to re-balance these systems such that the blockchain itself is a trusted agent which both insurers and the insured can confide in. We propose a "Smart Auto Insurance" system that minimizes data sharing while simultaneously providing quality-of-life improvements to both sides. Furthermore, we use a simple game theoretical argument to show that the clients using such a system are disincentivized from behaving adversarially., Comment: 15 pages, 1 figure, 2 tables

Information quality is of paramount importance to science. Accurate, scientifically vetted and statistically meaningful and, ideally, reproducible information engenders scientific trust and research opportunities. Not surprisingly, federal bodies (e.g., NASA, NOAA, USGS) have very strictly affirmed the importance of information quality in their product requirements. So-called Highly Influential Scientific Assessments (HISA) such as The Third US National Climate Assessment (NCA3) published in 2014 undergo a very rigorous review process to ensure transparency and credibility. To support the transparency of such reports, the U.S. Global Change Research Program (USGCRP) has developed the Global Change Information System (GCIS). A recent activity was performed to trace the provenance as completely as possible for all NCA3 figures that were predominantly based on NASA data. This poster presents the mechanics of that project and the lessons learned from that activity.

Earth Science informatics involves collaboration between multiple groups of people with diverse specializations and goals,often using variations in terminology to refer to common resources. The uniformity of the resource identifiers often does not cross organizational boundaries. Because of this, permanent, widely used, unambiguous identifiers for resources are elusive. We examine real world cases of changing and inconsistent identifiers which inherently work against persistence and uniformity. We also present a solution which mediates factors in these situations; namely the creation of lexicons:mappings of sets of terms to URIs which are curated within the Global Change Information System (GCIS). We discuss aspects of the GCIS which facilitate the use of lexicons: an information model which disambiguates resources, a RESTful API which provides metadata through content-negotiation, and a strategy for long term curation of URIs, including mechanisms for handling changes to URIs and variations in terms used by different communities while providing persistent URIs and preserving relationships between resources We provide working definitions of terms,contexts, and lexicons, and relate them to the practical challenges of disambiguation and curation. We also discuss the mechanisms employed and architecture of the GCIS, and how these choices facilitate representation of persistent identifiers and mappings of them to identifiers used colloquially within various earth science communities of practice.