From CourseWiki
Jump to: navigation, search

Main Page

CS 490 :: Search Engines & Recommendation Systems :: Spring 2017



Readings from Manning, Raghavan, & Schutze. Introduction to Information Retrieval (2008) unless otherwise stated.

Week Topic / Lecture Notes Readings Labs
Week 1 - 1/23 Intro, Boolean Retrieval Ch. 1, 2 Lab Overview, Survey
Week 2 - 1/30 Text Pre-Processing, Web Crawling Ch. 19, 20
RegEx Tutorial
in Python
Lab 1: Text Pre-Processing
Week 3 - 2/6 Designing a Meta-Search Engine Ch. 4.3 Lab 2: Web Crawling
Week 4 - 2/13 Ranked Retrieval Ch. 6,
TF-IDF Exercies
Lab 3: Boolean Retrieval
Week 5 - 2/20 Faster TF-IDF Ch. 6 Lab 4: Ranked Retrieval
Week 6 - 2/27 More Ranked-Retrieval Ch. 7 (optional)
Week 7 - 3/6 Evaluation Ch. 8 Lab 5: Evaluation
SB - 3/13 Spring Break IR Presentation Research Project
Week 8 - 3/20 Recommender Systems:
Intro (Doug, Wednesday)
Collaborative Filtering (David D., Eric G., Friday)

Ricci Ch 1
Netflix 2009 & Amazon 2003
Idea Generation
Week 9 - 3/27 Search Engines:
PageRank (Matt B., Chris W., Wednesday)
Architecture (David M., Will S., Friday)

Ch 21.1-21.2 & PageRank 1998
Google 1998
Literature Review
(Google Scholar)
Proposal Due (3/31)
Week 10 - 4/4 Classification:
naive Bayes (The Joes, Monday
kNN (Erika R. , Kizito U., Wednesday)

Ch. 13 & Better NB 2003
Ch. 14
Data Collection
(API, Crawl)
Week 11 - 4/10 Clustering:
Flat (Ryan D., Jimmy W., Monday)
Hierarchical (Noah Z., Yaw A., Wednesday)

Ch. 16
Ch. 17
(Search, Classify,
Cluster, Recommend)
Update Due (4/14)
Week 12 - 4/17 Recommendation:
Content-based (Shelby C., Jot S., Monday)
Music (Jeremy S., Trevor W., Wednesday)

CB 2007
Celma 2009 Ch 2 & 3
Week 13 - 4/24 Human Computation (Nicole L., Monday)
Final Exam (Wednesday)
ESPGame 2004 & reCaptcha 2008
Clean Up & Rerun
Week 14 - 5/1 NLP (Luke)
Final Project Presentations
Talk Slides Due
Finals Week (5/8) Final Project Reports Due
(Monday 5/8 at 5pm)

Course Overview

Search engines, such as Google, YouTube and Flickr, have had a huge impact on how people find and use information (e.g., webpages, videos, photos). Recommendation system like Netflix, Facebook, and Pandora, help people discover new and exciting things (e.g., movies, friends, songs). In this course, we will explore how information retrieval (IR) and recommendation systems (RecSys) are designed and implemented.

The first half of the class will be devoted to developing traditional IR skills such as web-crawling, text & multimedia processing, boolean & vector-space modeling, classification, clustering, and similarity analysis. The second half of the course will be devoted to creating a information retrieval or recommendation system as a collaborative class project. For this project, groups of students will design and develop individual components of this large-scale system. In the final weeks we will combine these components and (if all goes well) launch a new IR/RecSys for public use on the Internet.

COURSE FORMAT/STYLE: Lecture, Lab Meeting, Programming Assignments, Research Paper Reading and Dissection Collaborative Final Project

COURSE REQUIREMENTS & GRADING: Strong programming experience (CS220 or above) is required. Experience with Python is recommended. Advanced web programming (e.g., CS205) experience will also be useful but is not necessary.

Course Information

  • Prof. Doug Turnbull (, (607) 274-5743)
  • Class: MWF 12-12:50pm
  • Lab: Tu 2:35-3:50pm
  • Office Hours: Williams 321E
    • 2pm on Wednesdays, 3pm on Fridays in Williams 321E
    • By Appointment
    • Whenever my door is open
  • Course Teaching Assistant: Luke Waldner
    • Evening Help Sessions: 7-9pm on Wednesdays in Williams 309

Course Material

Related Courses