Santa Clara University

Mathematics and Computer Science department
 

Colloquium Series

Fall 2013

Unless otherwise noted, talks will be at 3:50 PM in O'Connor 107.  Also, there will be refreshments before each talk in O'Connor 31 at 3:30 PM.



Tuesday, October 15th

Speaker: Jeff Hamrick, University of San Francisco

Title: Are Umpires Racist?

Abstract: 

We investigate the racial preferences of Major League Baseball umpires as they evaluate both pitchers and hitters from 1989-2010. We find limited, and sometimes contradictory, evidence that umpires unduly favor or unjustly discriminate against players based on their race. Variables including attendance, terminal pitch, the absolute score differential, and the presence of monitoring systems do not consistently interact with umpire/pitcher and umpire/hitter racial combinations. Most evidence that would first appear to support racially-connected behaviors by umpires vanishes in three-way interaction models. Overall, in contrast with some of the other literature on this subject, our findings fall well short of convincing evidence for racial bias.

 


Tuesday, November 5th

Speaker: Angela Hicks, Stanford

Title: The Capricious Wives and the Diagonal Harmonics

Abstract: 

In 1966, Konheim and Weiss told the mathematical story of n dutiful husbands driving down a one way street and parking in the first available space upon receiving the command from their n (independently) capricious wives. We will discuss the now famous combinatorial object that results from the story--the parking function--and a few of the reasons for its study. In particular, we'll discuss a space of multivariate polynomials called the diagonal harmonics and their conjectured connection to the parking functions. We'll discuss open problems in the area, and time permitting, a connection to the Catalan numbers. This talk will assume basic familiarity with partial derivatives and some familiarity with linear algebra (especially the concept of dimension of a space) but no deeper background will be assumed.


 


Tuesday, November 12th

Speaker: Frank Farris, Santa Clara University

Title:  Polyhedral Symmetry in the Plane?

*** A Pi Mu Epsilon Sponsored Event

Abstract:  

When we classify plane patterns by their symmetries, there is a famous trichotomy: Plane patterns may be rosettes, friezes, or wallpaper patterns. The symmetries of a rosette all share a single fixed point; a frieze pattern is invariant under translation in one direction, a wallpaper pattern in two. In this talk, we undercut tradition, which normally insists that symmetries must preserve distances. We allow certain distance-deforming transformations to play the role of symmetries. In particular, we show how the polyhedral groups can act on the plane. To make patterns with these new transformations as symmetries, we construct functions invariant under the polyhedral actions. One of these is shown below. Do you believe that it has the same symmetries as a tetrahedron? This talk, accessible to undergraduate mathematics students, combines a little group theory, a little complex analysis, and several other ingredients in the service of mathematics and art.




FFarrisPic




 


Tuesday, December 3rd

Speaker: David Freeman, LinkedIn

Title:  Finding Spammy Names in Social Networks

Abstract:  Many social networks are predicated on the assumption that a member's online information reflects his or her real identity.  In such networks, members who fill their name fields with fictitious identities, company names, phone numbers, or just gibberish are violating the terms of service, polluting search results, and degrading the value of the site to real members.  Finding and removing these accounts on the basis of their spammy names can both improve the site experience for real members and prevent further abusive activity. 


In this talk we describe how to use the Naive Bayes classification algorithm to find accounts whose names do not represent real people.  The model can detect both automated and human abusers and can be used at registration time, before other signals such as social graph or clickstream history are present.  We use member data from LinkedIn to train and validate our model and
to choose parameters. 


We ran the algorithm on live LinkedIn data for one month in parallel with our previous name scoring algorithm based on regular expressions.  The false positive rate of our new algorithm (3.3%) was less than half that of the previous algorithm (7.0%).

 

 


If you have a disability and require a reasonable accommodation,
please call/email Rick Scott 408-554-4460/rscott at scu dot edu (or
use 1-800-735-2929 TTY—California Relay).

Printer-friendly format