BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Women@CL Talklet Event - Krittika D'Sliva\, Xiao Zhou\, Sian Goodi
 ng
DTSTART:20190215T130000Z
DTEND:20190215T140000Z
UID:TALK120322@talks.cam.ac.uk
CONTACT:Ayat Fekry
DESCRIPTION:*Speaker:* Krittika D'Sliva\n\n*Title:* Modeling Urban Environ
 ments with Complex Networks\n\n*Abstract* \nUrban environments have become
  a popular application in network science because of their complex dimensi
 onality\, from individuals moving from venue to venue to neighborhood conn
 ectivity changing over time. Better understanding cities is valuable as it
  enables us to model and predict changes in issues such as congestion\, en
 ergy consumption\, and environmental destruction. My work focuses on harne
 ssing traditional machine learning algorithms and network metrics to revea
 l properties of cities around the world. I will describe two recent works.
  In the first\, we utilize trends from temporally similar areas in a city 
 as features to predict the demand of a new venue. In the second\, we demon
 strate how a number of network and transport features can enable us to pre
 dict whether a business will survive or fail.\n\n----------------\n\n*Spea
 ker:* Xiao Zhou\n\n*Title:*Discovering Latent Patterns of Urban Cultural I
 nteractions in WeChat for Modern City Planning\n\n*Abstract*\nThe optimal 
 allocation of cultural establishments and related resources across urban r
 egions becomes of vital importance\, as it can reduce financial costs in t
 erms of planning and improve quality of life in the city\, more generally.
  We make use of a large longitudinal dataset of user location check-ins fr
 om the online social network WeChat to develop a data-driven framework for
  cultural planning in the city. We exploit rich spatio-temporal representa
 tions on user activity at cultural venues and use a novel extended version
  of the traditional latent Dirichlet allocation model that incorporates te
 mporal information to identify latent patterns of urban cultural interacti
 ons. Using the characteristic typologies of mobile user cultural activitie
 s emitted by the model\, we determine the levels of demand for different t
 ypes of cultural resources across urban areas. We then compare those with 
 the corresponding levels of supply as driven by the presence and spatial r
 each of cultural venues in local areas to obtain high resolution maps that
  indicate urban regions with lack of cultural resources\, and thus give su
 ggestions for further urban cultural planning and investment optimisation.
 \n\n----------------\n\n*Speaker:*Sian Gooding\n\n*Title:* Contextual Text
  Simplification\n\n*Abstract*\nTextual simplification can be defined as an
 y process that reduces the syntactical or lexical complexity of a text\, w
 hile attempting to preserve its meaning and information content. In this t
 alk I will explain the pros and cons of current state of the art deep lear
 ning simplification techniques. Additionally\, I shall present joint work 
 with Dr. Ekaterina Kochmar on contextual lexical simplification.\n\n------
 ----------\n\n
LOCATION:Computer Laboratory\, William Gates Building\, Room FW11
END:VEVENT
END:VCALENDAR