BEGIN:VCALENDAR
CALSCALE:GREGORIAN
VERSION:2.0
METHOD:PUBLISH
PRODID:-//Drupal iCal API//EN
X-WR-TIMEZONE:America/New_York
BEGIN:VTIMEZONE
TZID:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
DTSTART:20070311T020000
TZNAME:EDT
TZOFFSETTO:-0400
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
DTSTART:20071104T020000
TZNAME:EST
TZOFFSETTO:-0500
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
SEQUENCE:1
X-APPLE-TRAVEL-ADVISORY-BEHAVIOR:AUTOMATIC
UID:204306
DTSTAMP:20250213T095715Z
DTSTART;TZID=America/New_York:20250217T113000
DTEND;TZID=America/New_York:20250217T123000
URL;TYPE=URI:https://www.wpi.edu/news/calendar/events/computer-science-depa
 rtment-phd-research-qualifier-qian-wang-bridging-technical-analysis-and
SUMMARY:Computer Science Department, PhD Research Qualifier. Qian Wang "Bri
 dging Technical Analysis and Scientometric Insights in Biomedical Funding 
 with HGNNs Enhanced Calibration"
DESCRIPTION:\nQian Wang\nPhD Student\nWPI – Computer Science\n\nMonday, Feb
 ruary 17, 2025\nTime: 11:30 am – 12:30 pm\nZoom: https://wpi.zoom.us/j/522
 2416480\n\nCommittee Members :\nAdvisor: Prof. Xiaozhong Liu\nReader: Prof
 . Xiangnan Kong\nAbstract :\nThe concentration of research funding in a sm
 all segment of investigators and institutions (i.e., “the biomedical elite
 ”) has been well established. Moreover, funding inequality persists despit
 e deliberate efforts by funding agencies to counter it. To gain a deeper u
 nderstanding of how social capital might be driving this phenomena, we col
 lected publicly available data on the transition of over 11,000 National I
 nstitutes of Health Mentored Career Awardees (K01, K08, K23) from time of 
 their K award until their first R01-equivalent award (if any). Using data 
 from PubMed and other publicly available sources, we constructed a “hetero
 geneous scholarly graph” to represent the time-varying relationship betwee
 n MK awardees, the quality and quantity of their work over their careers, 
 their social capital (ties to influential people and institutions), and th
 eir ultimate success in obtaining R01-equivalent funding. We formulated an
 d tested predictors of ’K to R’ success in a graph neural network (GNN) mo
 del. In this paper we describe a novel process for calibrating a GNN model
  to improve its predictive accuracy called Heterogeneous Graph Calibration
  GNN (HGCGNN)– i.e., to align model predictions with observed outcomes. Af
 ter assigning a measure of quality to each node (i.e., the scholarly objec
 ts of interest, such published articles, journals, institutional affiliati
 ons, coauthors), we derived a feature subgraph for each node. Next, the qu
 ality and subgraphs of all neighboring nodes were concatenated to the targ
 et node. In order to calibrate the prediction of the ’K to R’ existence.\n
 Additionally, in the subgraph construction phase, we considered the influe
 nce of highly diverse neighbors on quality, calculating ‘K to R’ predictio
 n accuracy and node subgraph feature uniqueness to enhance subgraph repres
 entation. Our model simultaneously maintains accuracy and data efficiency.
 . We conducted empirical experiments to validate the effectiveness of our 
 model, demonstrating its consistent achievement of state-of-the-art calibr
 ation results across various graph datasets under different GNN backbones.
  Thus, the GNN confidence calibration improved the accuracy of our K to R 
 prediction model. This will facilitate research to better understand the r
 ole social capital plays in the distribution of NIH funding.\n
END:VEVENT
END:VCALENDAR