23 research outputs found
Profile of author and Portland native Heidi Julavits, featuring an interview in
No full textProfile of author and Portland native Heidi Julavits, featuring an interview in question-and-answer format. Julavits\u27s third novel, The Uses of Enchantment, is about a 17-year-old girl in West Salem, Mass., who may or may not have faked her own abduction. In addition to fiction writing, Julavits is founding editor of the Believer magazine. Julavits splits her time between Manhattan and Brooklin, Maine
Context-Based Quotation Recommendation
No full textWhile composing a new document, anything from a news article to an email or essay, authors often utilize direct quotes from a variety of sources. Although an author may know what point they would like to make, selecting an appropriate quote for the specific context may be time-consuming and difficult. We therefore propose a novel context-aware quote recommendation system which utilizes the content an author has already written to generate a ranked list of quotable paragraphs and spans of tokens from a given source document.
We approach quote recommendation as a variant of open-domain question answering and adapt the state-of-the-art BERT-based methods from open-QA to our task. We conduct experiments on a collection of speech transcripts and associated news articles, evaluating models' paragraph ranking and span prediction performances. Our experiments confirm the strong performance of BERT-based methods on this task, which outperform bag-of-words and neural ranking baselines by more than 30% relative across all ranking metrics. Qualitative analyses show the difficulty of the paragraph and span recommendation tasks and confirm the quotability of the best BERT model's predictions, even if they are not the true selected quotes from the original news articles
Predicting News Coverage of Scientific Articles
No full textJournalists act as gatekeepers to the scientific world, controlling what information reaches the public eye and how it is presented. Analyzing the kinds of research that typically receive more media attention is vital to understanding issues such as the “science of science communication” (National Academies of Sciences, Engineering, and Medicine 2017), patterns of misinformation, and the “cycle of hype.” We track the coverage of 91,997 scientific articles published in 2016 across various disciplines, publishers, and news outlets using metadata and text data from a leading tracker of scientific coverage in social and traditional media, Altmetric. We approach the problem as one of ranking each day’s, or week’s, papers by their likely level of media attention, using the learning-to-rank model lambdaMART (Burges 2010). We find that ngram features from the title, abstract and press release significantly improve performance over the metadata features journal, publisher, and subjects
Source Attribution: Recovering the Press Releases Behind Health Science News
No full textWe explore the task of intrinsic source attribution: inferring which portions of a derived document were adapted from an unobserved source document. Specifically, we model the relationship between news articles and their press release sources using a dataset of 64,784 health science news articles and 23,068 press releases. We approach the problem at the sentence level and work with science journalism professors to develop a four point Likert scale describing the extent to which a news article sentence is derived from the content in the corresponding press release. Because manual annotation of news article - press release pairs is time-consuming, we turn to a mix of expert, non-expert, and heuristic-based annotation to label our dataset. After a small pilot study, which found that humans, when only able to view the text of the news article, struggle to identify which content is derived or not, we compare four different sentence regression models on the task. We find that modeling a sentence's context in the entire document is important, with the best performing model, a sequence regression model with BERT token representations, achieving a spearman's ρ of 0.49 and NDCG@1 of 0.60 on the expert-labeled test set. Examining the model's predictions, we find that it successfully identifies copied or closely paraphrased sentences in articles with a mix of derived and original content, but struggles to differentiate between loosely paraphrased and original sentences in articles with mostly original writing
Archaeological and biological examination of “The Mystery Wreck” (8MO143) off Vaca Key, Monroe County, Florida: A Report Submitted to the Florida Keys National Marine Sanctuary in Fulfillment of a NOAA Maritime Heritage Program Mini-grant
No full textDuring the summer of 2004, the Florida Bureau of Archaeological Research Underwater Archaeology team undertook a project to relocate, assess, and record thirteen of the shipwrecks of the 1733 Spanish Plate Fleet in the Florida Keys. One source of background information that they used was a commercially available videotape entitled “Galleon Hunter,” produced by Don Ferguson. Aside from the 1733 wrecksites, the video features another site, locally known as “the Mystery Galleon,” that was shown to Ferguson by local diver Stefan Sykora. Using location numbers supplied in Ferguson’s video, Roger Smith, Della Scott-Ireton, and Dave McCampbell relocated the site in Hawk Channel, off the city of Marathon. Later, the site further was examined by Smith, Jennifer McKinnon, and Jason Raupp, who made initial sketches, still photos, and video recordings.ReportSubmitte
Evaluating the Effectiveness of Efficient Neural Architecture Search for Sentence-Pair Tasks
No full textImproper Supplementation Habits of Folic Acid Intake by Hungarian Pregnant Women: Improper Recommendations
No full textBackground: Neural tube defects (NTDs) are some of the most common congenital anomalies. Proper folic acid supplementation is a dominant risk factor, which has been shown to decrease the incidence of NTDs. In Canada, the incidence of neuroblastoma has presented a considerable decrease of 60% as a result of enrichment cereal grain flours with synthetic folic acid. The aim of this study was to investigate the effect of folic acid intake by pregnant women on the incidence of NTDs and neuroblastoma. Methods: Regular folic acid intake has been recommended to pregnant women in Hungary since the eighties of the last century by health visitors eventually raking effect as an official protocol which had been released in 1997. During 2001, 2002 and 2003. folic acid intake habits of pregnant women were evaluated by health visitors, proving to be successful in collecting data front 95.06% of the pregnant women. The incidence of NTDs has been registered by the Hungarian National Centre of Epidemiology, Department of Human Genetics and Teratology. The Pediatric Cancer Registry provided the incidence of neuroblastoma in children. Results: Consistent findings revealed a regular intake of supplementary folic acid products by 68.71% of the pregnant women. Out of these. 93.13% of pregnant women who were taking folic acid, started the supplementation after their 7 weeks of pregnancies, a time designated as the completion period of the development of the neural tube. The dose of folic acid supplementation was evaluated as less than 5 mg/day in 84.75% of the pregnant women. In Hungary, the incidence of NTDs has remained constant, while the incidence of neuroblastoma has shown constant slight increase in spite of the introduction of folic acid supplementation in 1997. Conclusions: Based on our experience, folic acid supplementation was initiated after the recognition of pregnancy and its application in a dose of lower than 5 mg/day neither decreased the incidence of NTDs nor did it have an effect on the neuroblastoma incidence. It is implicated that proper folic acid supplementation, which is started front the conception. can be achieved only with the enrichment of cereal grain flours
Preface
No full textThe University of Pennsylvania Working Papers in Linguistics (PWPL) is an occasional series published by the Penn Linguistics Club. The series has included volumes of previously unpublished work, or work in progress, by linguists with an ongoing affiliation with the Department, as well as volumes of papers from NWAV and the Penn Linguistics Colloquium.
This volume contains selected papers from the 33rd Penn Linguistics Colloquium, held from March 27-29, 2009 in Philadelphia, PA at the University of Pennsylvania.
Alphabetic thanks go to Dimka Atanassov, Toni Cook, Ariel Diertani, Lauren Friedman, Josef Fruehwald, Kyle Gorman, Catherine Lai, Marielle Lerner, Laurel MacKenzie, Brittany MacLaughlin, and Lydia Rieck for help in editing, uploading, and general support.
Since Vol. 14.2, PWPL has been an internet-only publication. Since Vol. 13.2, PWPL has been published both in print and online gratis via ScholarlyCommons@Penn. Due to the large number of hits these online papers have received, and the time and expense of managing a back catalog of PWPL volumes, the editorial committee decided in 2008 to cease print publication in favor of wider-scale free online dissemination. Please continue citing PWPL papers or issues as you would a print journal article, though you may also provide the URL of the manuscript. An example is below:
Antonenko, Andrei. 2010. Puzzles of Russian Subjunctives. U. Penn Working Papers in Linguistics 16.1: Proceedings of PLC 33, ed. J.S. Stevens, 1-10. http://repository.upenn.edu/pwpl/vol16/iss1/2
Ultimately, the entire back catalog will be digitized and made available on ScholarlyCommons@Penn.
Publication in the University of Pennsylvania Working Papers in Linguistics (PWPL) does not preclude submission of papers elsewhere; copyright is retained by the author(s) of individual papers.
The PWPL editors can be contacted at: U. Penn Working Papers in Linguistics 619 Williams Hall, University of Pennsylvania Philadelphia, PA 19104–6305
[email protected] http://ling.upenn.edu/papers/pwpl.html
Jon Scott Stevens
Issue Edito
Preface
No full textThe University of Pennsylvania Working Papers in Linguistics (PWPL) is an occasional series published by the Penn Linguistics Club. The series has included volumes of previously unpublished work, or work in progress, by linguists with an ongoing affiliation with the Department, as well as volumes of papers from NWAV and the Penn Linguistics Colloquium.
This volume contains selected papers from the 33rd Penn Linguistics Colloquium, held from March 27-29, 2009 in Philadelphia, PA at the University of Pennsylvania.
Alphabetic thanks go to Dimka Atanassov, Toni Cook, Ariel Diertani, Lauren Friedman, Josef Fruehwald, Kyle Gorman, Catherine Lai, Marielle Lerner, Laurel MacKenzie, Brittany MacLaughlin, and Lydia Rieck for help in editing, uploading, and general support.
Since Vol. 14.2, PWPL has been an internet-only publication. Since Vol. 13.2, PWPL has been published both in print and online gratis via ScholarlyCommons@Penn. Due to the large number of hits these online papers have received, and the time and expense of managing a back catalog of PWPL volumes, the editorial committee decided in 2008 to cease print publication in favor of wider-scale free online dissemination. Please continue citing PWPL papers or issues as you would a print journal article, though you may also provide the URL of the manuscript. An example is below:
Antonenko, Andrei. 2010. Puzzles of Russian Subjunctives. U. Penn Working Papers in Linguistics 16.1: Proceedings of PLC 33, ed. J.S. Stevens, 1-10. http://repository.upenn.edu/pwpl/vol16/iss1/2
Ultimately, the entire back catalog will be digitized and made available on ScholarlyCommons@Penn.
Publication in the University of Pennsylvania Working Papers in Linguistics (PWPL) does not preclude submission of papers elsewhere; copyright is retained by the author(s) of individual papers.
The PWPL editors can be contacted at: U. Penn Working Papers in Linguistics 619 Williams Hall, University of Pennsylvania Philadelphia, PA 19104–6305
[email protected] http://ling.upenn.edu/papers/pwpl.html
Jon Scott Stevens
Issue Edito
