BEGIN:VCALENDAR PRODID:-//Microsoft Corporation//Outlook 16.0 MIMEDIR//EN VERSION:2.0 METHOD:PUBLISH X-MS-OLK-FORCEINSPECTOROPEN:TRUE BEGIN:VTIMEZONE TZID:Central Standard Time BEGIN:STANDARD DTSTART:16011104T020000 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZOFFSETFROM:-0500 TZOFFSETTO:-0600 END:STANDARD BEGIN:DAYLIGHT DTSTART:16010311T020000 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZOFFSETFROM:-0600 TZOFFSETTO:-0500 END:DAYLIGHT END:VTIMEZONE BEGIN:VEVENT CLASS:PUBLIC CREATED:20230118T230601Z DESCRIPTION:Speakers:\nEmilio Lehoucq - PhD Candidate\, D epartment of Sociology\, Northwestern University\n\nAbby Smith - PhD Candidate\, Department of Statistics and Data Science \, Northwestern University\n\nEmma Zajdela - PhD Candidate\, Department of Engineering Sciences and A pplied Mathematics\, Northwestern University\n\nTitles and Abstracts:\nEmi lio Lehoucq - Do Americans Think the Digital Economy is Fair? Using Superv ised Learning to Explore Evaluations of Predictive Automation\nPredictive automation is a pervasive and archetypical example of the digital economy. Studying how Americans evaluate predictive automation is important becaus e it affects corporate and state governance. However\, we have relevant qu estions unanswered. We lack comparisons across use cases using a nationall y representative sample. We also have yet to determine what are the key pr edictors of evaluations of predictive automation. This article uses the Am erican Trends Panel’s 2018 wave (n=4\,594) to study whether American adu lts think predictive automation is fair across four use cases: helping cre dit decisions\, assisting parole decisions\, filtering job applicants base d on interview videos\, and assessing job candidates based on resumes. Res ults from lasso regressions trained with 112 predictors reveal that people ’s evaluations of predictive automation align with their views about soc ial media\, technology\, and politics.\nAbby Smith - The Impact of Entity Resolution on Observed Social Network Structure\nDeduplication\, also refe rred to as "entity resolution"\, is a common and crucial pre-processing st ep in the construction of social networks. Traditional deduplication metho ds compare the attributes (such as name and age) of potential matching pai rs to estimate a match probability for a pair. In social network datasets\ , we can also use relational information (e.g.\, a person’s network ties ) in deduplication. My work is focused on methods for evaluating entity re solution in a network setting\, measuring the sensitivity of entity resolu tion results to choices in tuning parameters\, and the downstream impacts these parameter choices can have on network metrics and topologies such as degree\, closeness\, and connectivity. I apply the evaluation methods to two real-world ego-centric network studies\, (i) Care2Hope\, a respondent- driven sample of rural people who use drugs (PWUD) in Appalachian Kentucky \, and (ii) RADAR\, a longitudinal network study of young men in Chicago w ho have sex with men.\nEmma Zajdela - Back in Fashion - Modeling the Cycli cal Dynamics of Trends \nCommon wisdom holds that fashion is cyclical\, wi th talk of trends coming "back\," from bell-bottom jeans to miniskirts. Hi storically\, a lack of quantitative data posed a barrier to explicit mathe matical study of this system\, however\, newly digitized historical record s now make such work possible. This talk will present analysis from a new database we constructed quantifying tens of thousands of images of clothin g from 1869 to present day. It will describe approaches to modeling the cy clical dynamics of fashion observed in the dataset. Large-scale social phe nomena such as fashion trends are of intrinsic interest themselves\, but c an also help elucidate the interplay of creativity\, differentiation\, con formity\, and diffusion of ideas in broader human systems. Acknowledgement s: This work was supported by the NSF Graduate Research Fellowship and the NICO Intersection Science Fellowship. \nLocation:\nIn person: Chambers Ha ll \, 600 Foster Street\, Lower Level\nRemote option: https://northwestern.zoom.us/j/92972622141 \nPasscode: NICO23\nAbout the Speaker Series:\nWednesdays@NICO is a v ibrant weekly seminar series focusing broadly on the topics of complex sys tems and data science. It brings together attendees ranging from graduate students to senior faculty who span all of the schools across Northwestern \, from applied math to sociology to biology and every discipline in-betwe en. Please visit: https://bit.ly/WedatNICO for information on future speak ers.\n \n \n DTEND;TZID="Central Standard Time":20230201T130000 DTSTAMP:20230118T230601Z DTSTART;TZID="Central Standard Time":20230201T120000 LAST-MODIFIED:20230118T230601Z LOCATION:In person at Chambers Hall\, or remote via Zoom PRIORITY:5 SEQUENCE:0 SUMMARY;LANGUAGE=en-us:Wed@NICO\, 2/1\, Lightning Talks with NU Scholars TRANSP:OPAQUE UID:040000008200E00074C5B7101A82E0080000000050749A3A4C2BD901000000000000000 010000000DDC1F01433679C48922D83B205066152 X-ALT-DESC;FMTTYPE=text/html:

Speakers:

Emilio Lehoucq - PhD Candi date\, Department of Sociology\, Northwestern University

Abby Smith - PhD Candidate\, Department of Stat istics and Data Science\, Northwestern University

Emma Zajdela - PhD Candidate\, Department of Engineering Sciences and Applied Mat hematics\, Northwestern University

Titles and Abstra cts:

Emilio Lehoucq - Do Americans Think the Digital Economy is Fair? Using Supervised Learning to Explore Evaluations of Predictive Automation< /o:p>

Predictive automation is a pervasive and archetypical example of the digital economy. Studying how Americans evalua te predictive automation is important because it affects corporate and sta te governance. However\, we have relevant questions unanswered. We lack co mparisons across use cases using a nationally representative sample. We al so have yet to determine what are the key predictors of evaluations of pre dictive automation. This article uses the American Trends Panel’\;s 2 018 wave (n=4\,594) to study whether American adults think predictive auto mation is fair across four use cases: helping credit decisions\, assisting parole decisions\, filtering job applicants based on interview videos\, a nd assessing job candidates based on resumes. Results from lasso regressio ns trained with 112 predictors reveal that people’\;s evaluations of predictive automation align with their views about social media\, technolo gy\, and politics.

Abby Smith - The Impact of Entity Resolution on Observed Social N etwork Structure

Deduplication\, als o referred to as "\;entity resolution"\;\, is a common and crucial pre-processing step in the construction of social networks. Traditional d eduplication methods compare the attributes (such as name and age) of pote ntial matching pairs to estimate a match probability for a pair. In social network datasets\, we can also use relational information (e.g.\, a perso n’\;s network ties) in deduplication. My work is focused on methods f or evaluating entity resolution in a network setting\, measuring the sensi tivity of entity resolution results to choices in tuning parameters\, and the downstream impacts these parameter choices can have on network metrics and topologies such as degree\, closeness\, and connectivity. I apply the evaluation methods to two real-world ego-centric network studies\, (i) Ca re2Hope\, a respondent-driven sample of rural people who use drugs (PWUD) in Appalachian Kentucky\, and (ii) RADAR\, a longitudinal network study of young men in Chicago who have sex with men.

Emma Zajdela - Back in Fashion - Modeling the Cyclical Dynamics of Trends \;

Common wisdom holds that fashion is cyclic al\, with talk of trends coming "\;back\,"\; from bell-bottom jean s to miniskirts. Historically\, a lack of quantitative data posed a barrie r to explicit mathematical study of this system\, however\, newly digitize d historical records now make such work possible. This talk will present a nalysis from a new database we constructed quantifying tens of thousands o f images of clothing from 1869 to present day. It will describe approaches to modeling the cyclical dynamics of fashion observed in the dataset. Lar ge-scale social phenomena such as fashion trends are of intrinsic interest themselves\, but can also help elucidate the interplay of creativity\, di fferentiation\, conformity\, and diffusion of ideas in broader human syste ms. Acknowledgements: This work was supported by the NSF Graduate Research Fellowship and the NICO Intersection Science Fellowship. \;

Loc ation:

In person: Chambers Hall\, 600 Foster Street\, L ower Level
Remote option: https://northwestern.z oom.us/j/92972622141
Passcode: NICO23

About the Speaker Series:

Wednesdays@NICO is a vibrant weekly seminar series focusing broadly on the topics of complex systems and data science. It brings together attendees ranging from graduate students to senior fac ulty who span all of the schools across Northwestern\, from applied math t o sociology to biology and every discipline in-between. Please visit: https://bit.ly/WedatNICO for informatio n on future speakers.

 \;< /p>

 \;

X-MICROSOFT-CDO-BUSYSTATUS:BUSY X-MICROSOFT-CDO-IMPORTANCE:1 X-MICROSOFT-DISALLOW-COUNTER:FALSE X-MS-OLK-AUTOFILLLOCATION:FALSE X-MS-OLK-CONFTYPE:0 BEGIN:VALARM TRIGGER:-PT15M ACTION:DISPLAY DESCRIPTION:Reminder END:VALARM END:VEVENT END:VCALENDAR