BEGIN:VCALENDAR
PRODID:-//eluceo/ical//2.0/EN
VERSION:2.0
CALSCALE:GREGORIAN
BEGIN:VEVENT
UID:c1dcb7809901986c497013562bdd3ade
DTSTAMP:20240507T233025Z
SUMMARY:Dr. Xiao Fang\, University of Delaware
DESCRIPTION: \n\nTitle: A Deep Learning Approach to Industry Classification
 \n\nAbstract: Business researchers and practitioners often attempt to\ndiv
 ide firms into more homogeneous groups that bear similar\ncharacteristics.
  In numerous instances\, industry classification\nsystems (ICSs) have been
  selected to achieve this objective.\nTraditional expert-driven ICSs\, suc
 h as Standard Industrial\nClassification (SIC) and Global Industry Classif
 ication Standard\n(GICS)\, have several issues including limited granulari
 ty and high\ncost of maintenance. An alternative algorithm-driven approach
  to ICS\ninfers similarities between firms from their financial documents\
 n(e.g.\, annual 10-K reports). While the algorithm-driven approach is\npro
 mising in circumventing the limitations faced by the expert-driven\napproa
 ch\, one needs to solve the challenging problem of text\nrepresentation\, 
 that is\, how to accurately represent textual documents\nwith numerical fe
 atures. In this paper\, we propose a novel text\nrepresentation model base
 d on deep learning. The proposed model is\ncapable of handling arbitrary l
 ong documents with heterogeneous\nconcepts. It balances the trade-off betw
 een memorizing local details\nand capturing global concepts by introducing
  a multiplicative\nmechanism to integrate information from contexts of dif
 ferent scopes.\nWe further develop an ICS by applying the proposed text re
 presentation\nmodel to Item 1 and 1A sections of firms’ annual 10K repor
 ts. We\ndemonstrate the effectiveness of the proposed text representation\
 nmodel by benchmarking the developed ICS against state-of-the-art\nexpert-
 driven and algorithm-driven ICSs.\n\nLearn more about Deep Analytics [http
 ://dalab.info/index.html]\n\nRead more about Dr. Fang\n[https://www.udel.e
 du/faculty-staff/experts/xiao-fang/]\n 
DTSTART:20190531T143000Z
DTEND:20190531T160000Z
LOCATION:Gerri C. LeBow Hall\, 3220 Market Street\, 209\, Philadelphia\, PA
  19104
END:VEVENT
END:VCALENDAR