Distributed Application System 분산응용시스템

Dept. Electronics and Computer Engineering Chonnam National University ( English)

Description

The purpose of this graduate-level course is introducing a distributed application system and understanding the related issues. In this semester, the main topic of this course is a recommender system. The course covers the essential of recommender systems such as recommendation techniques (collaborative, content-based, knowledge-based and hybrid) and evaluation methods, as well as exploring some further issues such as security issues, online market decision making process, recommendation systems in next generation web or IoT/ubiquitous environment. Also, this course intend that students catch up the recent research issues related to recommender systems. Extensive paper reading/presentation assignments and a project related to recommender systems will be issued.

Instructor

Kyungbaek Kim
Office : Engineering Building #6, 715
Tel : +82-62-530-3438
Email : kyungbaekkim@jnu.ac.kr
Office Hours : Wed 10:00 ~ 10:45

Time and Location

Thur 9:00-11:30, Engineering Building #6, 106

Textbook

• Recommender Systems : An Introduction, by Dietmar Jnnach, Markus Zanker, Alexander Felfering, and Gerhard Friedrich

Grading Policy

• Attendance : 10%
• Reading Assignments and exercises : 30%
  - Tentatively Two papers per week : 13 papers
  - Hadoop exercises
  
• Projects : 30%
  - Personal Project : research on Hadoop related projects
  - Team Project
  
• Exam : 30%

Lecture Notes

Lecture notes are accessible through the eClass of JNU portal.

• Syllabus
• 1.Introduction
• 2.Collaborative recommendation
• 3.Content-based recommendation
• 4.Knowledge-based recommendation
• 5.Hybrid recommendation approaches

Homeworks, Quiz, Midterm/Final Exam

All of the materials related to homeworks, quiz, midterm exam and final exam, including solutions, are accessible through the eClass of JNU portal.

Exercise of Recommender System

• Exercise 01

Reading Assignment

Submit the summary of given reading assignments on the due date.

• Due on 11th March
  • Collaborative recommendation
• Due on 18th March
  • Content-based recommendation
• Due on 25th March
  • Knowledge based recommendation
• Due on 1st April
  • Hybrid recommendation approaches
  • Evaluating recommender systems
• Due on 8th April
  paper 25 and 26
• Due on 15th April
  paper 1,2, and 3
• Due on 29th April
  paper 4,5, and 6
• Due on 6th May
  paper 7,8, and 9
• Due on 13th May
  paper 10,11, and 12
• Due on 20th May
  paper 13,14, and 15
• Due on 27th May
  paper 16,17, and 18
• Due on 3rd June
  paper 19,20, and 21
• Due on 10th June
  paper 22,23, and 24

Paper Presentation

1. [CIKM_2013] Location Recommendation for Out-of-Town Users in Location-Based Social Networks
2. [CIKM_2013] GAPfm Optimal Top-N Recommendations for Graded Relevance Domains
3. [CIKM_2013] Community-Based User Recommendation in Uni-Directional Social Networks
4. [ICDE_2013] Focused Matrix Factorization For Audience Selection in Display Advertising
5. [KDD_2013] FISM factored item similarity models for top-N recommender systems
6. [KDD_2013] LCARS a location-content-aware recommender system
7. [KDD_2013] Learning geographical preferences for point-of-interest recommendation
8. [WWW_2014] A Monte Carlo Algorithm for Cold Start Recommendation
9. [WWW_2014] Local collabarative ranking
10. [WWW_2014] Personalized Collaborative Clustering
11. [WWW_2014] Temporal QoS-Aware Web Service Recommendation via Non-negative Tensor
12. [CIKM_2014] Graph-based Point-of-interest Recommendation with Geographical and Temporal Influences
13. [CIKM_2014] Meta-Path-Based Ranking with Pseudo Relevance Feedback on Heterogeneous Graph for Citation Recommendation
14. [CIKM_2014] On Improving Co-Cluster Quality with Application to Recommender Systems
15. [CIKM_2014] User Interests Imbalance Exploration in Social Recommendation- A Fitness Adaptation
16. [ICDE_2014] CrowdPlanner- A Crowd-Based Route Recommendation System
17. [ICDE_2014] Efficient Instant-Fuzzy Search with Proximity Ranking
18. [ICDE_2014] Exploiting group recommendation functions for flexible preferences
19. [ICDE_2014] Ranking Item Features by Mining Online User-Item Interactions
20. [KDD_2014] Jointly Modeling Aspects, Ratings and Sentiments for Movie Recommendation (JMARS)
21. [KDD_2014] Matching Users and Items Across Domains to Improve the Recommendation Quality
22. [KDD_2014] Optimal Recommendations under Attraction, Aversion, and Social Influence
23. [KDD_2014] GeoMF- Joint Geographical Modeling and Matrix Factorization for Point-of-Interest Recommendation
24. [KDD_2014] Product Selection Problem- Improve Market Share by Learning Consumer Behavior
25. [SIGIR_2013] Time-aware Point-of-interest Recommendation
26. [SIGIR 2014] Addressing Cold Start in Recommender Systems A Semi-supervised Co-training Algorithm
27. [SIGIR 2014] Explicit Factor Models for Explainable Recommendation based on Phrase-level Sentiment Analysis
28. [SIGIR 2014] Gaussian Process Factorization Machines for Context-aware Recommendations

Date	Student	Paper	Slides
2015-April-09	정우주 Duong	25 26	slide slide
2015-April-16	Zubair Afaq Ngoc	1 2 3	slide slide slide
2015-April-30	Gde Fiqri Nhat	4 5 6	slide slide slide
2015-May-07	Rajashree Priagung Alvin	7 8 9	slide slide slide
2015-May-14	Leseley Gemoh Tiep	19 11 12	slide slide slide
2015-June-04	Zubair Afaq Fiqri Nhat Priagung Alvin	13 14 15 16 17 18	slide slide slide slide slide slide
2015-June-11	Lesley Gemoh Gde Tiep Rajashree Ngoc Duong	10 20 21 22 23 24 28	slide slide slide slide slide slide slide

Project

Team #	Members	Title
1	Rajashree, Tiep, Lesley, Gemoh	Graph-based interested user recommendation by using geo-social data
2	Nhat, Ngoc, Duong	Solving Sparsity Challenges in Collaborative Filtering System
3	Fiqri, Alvin, Priagung, Gde	Soft-thresholded SVD Recommender Application Using "softImpute" library on R
4	Zubair, Afaq	Summary on hybrid recommendation approaches

Last Modified by Kyungbaek Kim, June 11 2015.