arXiv subjects

Shawn M. Jones

Publications and source records attributed to Shawn M. Jones.

16 recordsLinked to original sources

arXiv2025

LLM-Assisted Translation of Legacy FORTRAN Codes to C++: A Cross-Platform Study

Large Language Models (LLMs) are increasingly being leveraged for generating and translating scientific computer codes by both domain-experts and non-domain experts. Fortran has served as one of the go to programming languages in legacy high-performance computing (HPC) for scientific discoveries. Despite growing adoption, LLM-based code translation of legacy code-bases has not been thoroughly assessed or quantified for its usability. Here, we studied the applicability of LLM-based translation of Fortran to C++ as a step towards building an agentic-workflow using open-weight LLMs on two different computational platforms. We statistically quantified the compilation accuracy of the translated C++ codes, measured the similarity of the LLM translated code to the human translated C++ code, and statistically quantified the output similarity of the Fortran to C++ translation.

cs.SE↗

arXiv2023

DeepPatent2: A Large-Scale Benchmarking Corpus for Technical Drawing Understanding

Recent advances in computer vision (CV) and natural language processing have been driven by exploiting big data on practical applications. However, these research fields are still limited by the sheer volume, versatility, and diversity of the available datasets. CV tasks, such as image captioning, which has primarily been carried out on natural images, still struggle to produce accurate and meaningful captions on sketched images often included in scientific and technical documents. The advancement of other tasks such as 3D reconstruction from 2D images requires larger datasets with multiple viewpoints. We introduce DeepPatent2, a large-scale dataset, providing more than 2.7 million technical drawings with 132,890 object names and 22,394 viewpoints extracted from 14 years of US design patent documents. We demonstrate the usefulness of DeepPatent2 with conceptual captioning. We further provide the potential usefulness of our dataset to facilitate other research areas such as 3D image reconstruction and image retrieval.

cs.CV↗

arXiv2023

Discovering Image Usage Online: A Case Study With "Flatten the Curve''

Understanding the spread of images across the web helps us understand the reuse of scientific visualizations and their relationship with the public. The "Flatten the Curve" graphic was heavily used during the COVID-19 pandemic to convey a complex concept in a simple form. It displays two curves comparing the impact on case loads for medical facilities if the populace either adopts or fails to adopt protective measures during a pandemic. We use five variants of the "Flatten the Curve" image as a case study for viewing the spread of an image online. To evaluate its spread, we leverage three information channels: reverse image search engines, social media, and web archives. Reverse image searches give us a current view into image reuse. Social media helps us understand a variant's popularity over time. Web archives help us see when it was preserved, highlighting a view of popularity for future researchers. Our case study leverages document URLs can be used as a proxy for images when studying the spread of images online.

cs.SI↗

arXiv2022

Abstract Images Have Different Levels of Retrievability Per Reverse Image Search Engine

Much computer vision research has focused on natural images, but technical documents typically consist of abstract images, such as charts, drawings, diagrams, and schematics. How well do general web search engines discover abstract images? Recent advancements in computer vision and machine learning have led to the rise of reverse image search engines. Where conventional search engines accept a text query and return a set of document results, including images, a reverse image search accepts an image as a query and returns a set of images as results. This paper evaluates how well common reverse image search engines discover abstract images. We conducted an experiment leveraging images from Wikimedia Commons, a website known to be well indexed by Baidu, Bing, Google, and Yandex. We measure how difficult an image is to find again (retrievability), what percentage of images returned are relevant (precision), and the average number of results a visitor must review before finding the submitted image (mean reciprocal rank). When trying to discover the same image again among similar images, Yandex performs best. When searching for pages containing a specific image, Google and Yandex outperform the others when discovering photographs with precision scores ranging from 0.8191 to 0.8297, respectively. In both of these cases, Google and Yandex perform better with natural images than with abstract ones achieving a difference in retrievability as high as 54\% between images in these categories. These results affect anyone applying common web search engines to search for technical documents that use abstract images.

cs.CV↗

arXiv2022

Creating Structure in Web Archives With Collections: Different Concepts From Web Archivists

As web archives' holdings grow, archivists subdivide them into collections so they are easier to understand and manage. In this work, we review the collection structures of eight web archive platforms: : Archive-It, Conifer, the Croatian Web Archive (HAW), the Internet Archive's user account web archives, Library of Congress (LC), PANDORA, Trove, and the UK Web Archive (UKWA). We note a plethora of different approaches to web archive collection structures. Some web archive collections support sub-collections and some permit embargoes. Curatorial decisions may be attributed to a single organization or many. Archived web pages are known by many names: mementos, copies, captures, or snapshots. Some platforms restrict a memento to a single collection and others allow mementos to cross collections. Knowledge of collection structures has implications for many different applications and users. Visitors will need to understand how to navigate collections. Future archivists will need to understand what options are available for designing collections. Platform designers need it to know what possibilities exist. The developers of tools that consume collections need to understand collection structures so they can meet the needs of their users.

cs.DL↗

arXiv2021

It's All About The Cards: Sharing on Social Media Probably Encouraged HTML Metadata Growth

In a perfect world, all articles consistently contain sufficient metadata to describe the resource. We know this is not the reality, so we are motivated to investigate the evolution of the metadata that is present when authors and publishers supply their own. Because applying metadata takes time, we recognize that each news article author has a limited metadata budget with which to spend their time and effort. How are they spending this budget? What are the top metadata categories in use? How did they grow over time? What purpose do they serve? We also recognize that not all metadata fields are used equally. What is the growth of individual fields over time? Which fields experienced the fastest adoption? In this paper, we review 227,726 HTML news articles from 29 outlets captured by the Internet Archive between 1998 and 2016. Upon reviewing the metadata fields in each article, we discovered that 2010 began a metadata renaissance as publishers embraced metadata for improved search engine ranking, search engine tracking, social media tracking, and social media sharing. When analyzing individual fields, we find that one application of metadata stands out above all others: social cards -- the cards generated by platforms like Twitter when one shares a URL. Once a metadata standard was established for cards in 2010, its fields were adopted by 20% of articles in the first year and reached more than 95% adoption by 2016. This rate of adoption surpasses efforts like Schema.org and Dublin Core by a fair margin. When confronted with these results on how news publishers spend their metadata budget, we must conclude that it is all about the cards.

cs.DL↗

arXiv2021

Automatically Selecting Striking Images for Social Cards

To allow previewing a web page, social media platforms have developed social cards: visualizations consisting of vital information about the underlying resource. At a minimum, social cards often include features such as the web resource's title, text summary, striking image, and domain name. News and scholarly articles on the web are frequently subject to social card creation when being shared on social media. However, we noticed that not all web resources offer sufficient metadata elements to enable appealing social cards. For example, the COVID-19 emergency has made it clear that scholarly articles, in particular, are at an aesthetic disadvantage in social media platforms when compared to their often more flashy disinformation rivals. Also, social cards are often not generated correctly for archived web resources, including pages that lack or predate standards for specifying striking images. With these observations, we are motivated to quantify the levels of inclusion of required metadata in web resources, its evolution over time for archived resources, and create and evaluate an algorithm to automatically select a striking image for social cards. We find that more than 40% of archived news articles sampled from the NEWSROOM dataset and 22% of scholarly articles sampled from the PubMed Central dataset fail to supply striking images. We demonstrate that we can automatically predict the striking image with a Precision@1 of 0.83 for news articles from NEWSROOM and 0.78 for scholarly articles from the open access journal PLOS ONE.

cs.DL↗

arXiv2020

MementoEmbed and Raintale for Web Archive Storytelling

For traditional library collections, archivists can select a representative sample from a collection and display it in a featured physical or digital library space. Web archive collections may consist of thousands of archived pages, or mementos. How should an archivist display this sample to drive visitors to their collection? Search engines and social media platforms often represent web pages as cards consisting of text snippets, titles, and images. Web storytelling is a popular method for grouping these cards in order to summarize a topic. Unfortunately, social media platforms are not archive-aware and fail to consistently create a good experience for mementos. They also allow no UI alterations for their cards. Thus, we created MementoEmbed to generate cards for individual mementos and Raintale for creating entire stories that archivists can export to a variety of formats.

cs.DL↗

arXiv2020

SHARI -- An Integration of Tools to Visualize the Story of the Day

Tools such as Google News and Flipboard exist to convey daily news, but what about the past? In this paper, we describe how to combine several existing tools with web archive holdings to perform news analysis and visualization of the "biggest story" for a given date. StoryGraph clusters news articles together to identify a common news story. Hypercane leverages ArchiveNow to store URLs produced by StoryGraph in web archives. Hypercane analyzes these URLs to identify the most common terms, entities, and highest quality images for social media storytelling. Raintale then uses the output of these tools to produce a visualization of the news story for a given day. We name this process SHARI (StoryGraph Hypercane ArchiveNow Raintale Integration).

cs.DL↗

arXiv2019

Social Cards Probably Provide For Better Understanding Of Web Archive Collections

Used by a variety of researchers, web archive collections have become invaluable sources of evidence. If a researcher is presented with a web archive collection that they did not create, how do they know what is inside so that they can use it for their own research? Search engine results and social media links are represented as surrogates, small easily digestible summaries of the underlying page. Search engines and social media have a different focus, and hence produce different surrogates than web archives. Search engine surrogates help a user answer the question "Will this link meet my information need?" Social media surrogates help a user decide "Should I click on this?" Our use case is subtly different. We hypothesize that groups of surrogates together are useful for summarizing a collection. We want to help users answer the question of "What does the underlying collection contain?" But which surrogate should we use? With Mechanical Turk participants, we evaluate six different surrogate types against each other. We find that the type of surrogate does not influence the time to complete the task we presented the participants. Of particular interest are social cards, surrogates typically found on social media, and browser thumbnails, screen captures of web pages rendered in a browser. At $p=0.0569$, and $p=0.0770$, respectively, we find that social cards and social cards paired side-by-side with browser thumbnails probably provide better collection understanding than the surrogates currently used by the popular Archive-It web archiving platform. We measure user interactions with each surrogate and find that users interact with social cards less than other types. The results of this study have implications for our web archive summarization work, live web curation platforms, social media, and more.

cs.DL↗

arXiv2018

The Off-Topic Memento Toolkit

Web archive collections are created with a particular purpose in mind. A curator selects seeds, or original resources, which are then captured by an archiving system and stored as archived web pages, or mementos. The systems that build web archive collections are often configured to revisit the same original resource multiple times. This is incredibly useful for understanding an unfolding news story or the evolution of an organization. Unfortunately, over time, some of these original resources can go off-topic and no longer suit the purpose for which the collection was originally created. They can go off-topic due to web site redesigns, changes in domain ownership, financial issues, hacking, technical problems, or because their content has moved on from the original topic. Even though they are off-topic, the archiving system will still capture them, thus it becomes imperative to anyone performing research on these collections to identify these off-topic mementos. Hence, we present the Off-Topic Memento Toolkit, which allows users to detect off-topic mementos within web archive collections. The mementos identified by this toolkit can then be separately removed from a collection or merely excluded from downstream analysis. The following similarity measures are available: byte count, word count, cosine similarity, Jaccard distance, Sørensen-Dice distance, Simhash using raw text content, Simhash using term frequency, and Latent Semantic Indexing via the gensim library. We document the implementation of each of these similarity measures. We possess a gold standard dataset generated by manual analysis, which contains both off-topic and on-topic mementos. Using this gold standard dataset, we establish a default threshold corresponding to the best F1 score for each measure. We also provide an overview of potential future directions that the toolkit may take.

cs.DL↗

arXiv2018

The Many Shapes of Archive-It

Web archives, a key area of digital preservation, meet the needs of journalists, social scientists, historians, and government organizations. The use cases for these groups often require that they guide the archiving process themselves, selecting their own original resources, or seeds, and creating their own web archive collections. We focus on the collections within Archive-It, a subscription service started by the Internet Archive in 2005 for the purpose of allowing organizations to create their own collections of archived web pages, or mementos. Understanding these collections could be done via their user-supplied metadata or via text analysis, but the metadata is applied inconsistently between collections and some Archive-It collections consist of hundreds of thousands of seeds, making it costly in terms of time to download each memento. Our work proposes using structural metadata as an additional way to understand these collections. We explore structural features currently existing in these collections that can unveil curation and crawling behaviors. We adapt the concept of the collection growth curve for understanding Archive-It collection curation and crawling behavior. We also introduce several seed features and come to an understanding of the diversity of resources that make up a collection. Finally, we use the descriptions of each collection to identify four semantic categories of Archive-It collections. Using the identified structural features, we reviewed the results of runs with 20 classifiers and are able to predict the semantic category of a collection using a Random Forest classifier with a weighted average F1 score of 0.720, thus bridging the structural to the descriptive. Our method is useful because it saves the researcher time and bandwidth. Identifying collections by their semantic category allows further downstream processing to be tailored to these categories.

cs.DL↗

arXiv2016

Persistent URIs Must Be Used To Be Persistent

We quantify the extent to which references to papers in scholarly literature use persistent HTTP URIs that leverage the Digital Object Identifier infrastructure. We find a significant number of references that do not, speculate why authors would use brittle URIs when persistent ones are available, and propose an approach to alleviate the problem.

cs.DL↗

arXiv2016

Rules of Acquisition for Mementos and Their Content

Text extraction from web pages has many applications, including web crawling optimization and document clustering. Though much has been written about the acquisition of content from live web pages, content acquisition of archived web pages, known as mementos, remains a relatively new enterprise. In the course of conducting a study with almost 700,000 web pages, we encountered issues acquiring mementos and extracting text from them. The acquisition of memento content via HTTP is expected to be a relatively painless exercise, but we have found cases to the contrary. We also find that the parsing of HTML, already known to be problematic, can be more complex when one attempts to extract the text of mementos across many web archives, due to issues involving different memento presentation behaviors, as well as the age of the HTML in their mementos. For the benefit of others acquiring mementos across many web archives, we document those experiences here.

cs.DL↗

arXiv2015

Avoiding Spoilers in Fan Wikis of Episodic Fiction

A variety of fan-based wikis about episodic fiction (e.g., television shows, novels, movies) exist on the World Wide Web. These wikis provide a wealth of information about complex stories, but if readers are behind in their viewing they run the risk of encountering "spoilers" -- information that gives away key plot points before the intended time of the show's writers. Enterprising readers might browse the wiki in a web archive so as to view the page prior to a specific episode date and thereby avoid spoilers. Unfortunately, due to how web archives choose the "best" page, it is still possible to see spoilers (especially in sparse archives). In this paper we discuss how to use Memento to avoid spoilers. Memento uses TimeGates to determine which best archived page to give back to the user, currently using a minimum distance heuristic. We quantify how this heuristic is inadequate for avoiding spoilers, analyzing data collected from fan wikis and the Internet Archive. We create an algorithm for calculating the probability of encountering a spoiler in a given wiki article. We conduct an experiment with 16 wiki sites for popular television shows. We find that 38% of those pages are unavailable in the Internet Archive. We find that when accessing fan wiki pages in the Internet Archive there is as much as a 66% chance of encountering a spoiler. Using sample access logs from the Internet Archive, we find that 19% of actual requests to the Wayback Machine for wikia.com pages ended in spoilers. We suggest the use of a different minimum distance heuristic, minpast, for wikis, using the desired datetime as an upper bound.

cs.DL↗

arXiv2014

Bringing Web Time Travel to MediaWiki: An Assessment of the Memento MediaWiki Extension

We have implemented the Memento MediaWiki Extension Version 2.0, which brings the Memento Protocol to MediaWiki, used by Wikipedia and the Wikimedia Foundation. Test results show that the extension has a negligible impact on performance. Two 302 status code datetime negotiation patterns, as defined by Memento, have been examined for the extension: Pattern 1.1, which requires 2 requests, versus Pattern 2.1, which requires 3 requests. Our test results and mathematical review find that, contrary to intuition, Pattern 2.1 performs better than Pattern 1.1 due to idiosyncrasies in MediaWiki. In addition to implementing Memento, Version 2.0 allows administrators to choose the optional 200-style datetime negotiation Pattern 1.2 instead of Pattern 2.1. It also permits administrators the ability to have the Memento MediaWiki Extension return full HTTP 400 and 500 status codes rather than using standard MediaWiki error pages. Finally, version 2.0 permits administrators to turn off recommended Memento headers if desired. Seeing as much of our work focuses on producing the correct revision of a wiki page in response to a user's datetime input, we also examine the problem of finding the correct revisions of the embedded resources, including images, stylesheets, and JavaScript; identifying the issues and discussing whether or not MediaWiki must be changed to support this functionality.

cs.DL↗