Automatic Advertisement Rating

Slides:



Advertisements
Similar presentations
Ke Chen 1, Shaogang Gong 1, Tao Xiang 1, Chen Change Loy 2 1. Queen Mary, University of London 2. The Chinese University of Hong Kong VGG reading group.
Advertisements

Chapter Fourteen Market Positioning and Branding.
A Framework for Photo-Quality Assessment and Enhancement based on Visual Aesthetics Subhabrata Bhattacharya Rahul Sukthankar Mubarak Shah.
Predicting Protein Interactions HERPES! Team Question Mark Jeff Brown Dante Kappotis Robert Vanderley Anthony Biasella.
Cause Marketing Do Consumers buy it? Who is taking advantage of it?
Weka solution for the 2004 KDD Cup Protein Homology Prediction task Bernhard Pfahringer Weka Group, University of Waikato, New Zealand.
WEEK 6: DEEP TRACKING STUDENTS: SI CHEN & MEERA HAHN MENTOR: AFSHIN DEGHAN.
Automatic Gender Identification using Cell Phone Calling Behavior Presented by David.
The basic idea So far, we have been comparing two samples
Question Identification on Twitter Baichuan Li, Xiance Si, Michael R. Lyu, Irwin King, and Edward Y. Chang 10/9/20151.
Business Management/ Entrepreneurship Class Field Trip 2012 All rights reserved Carol Arcomano 2012.
1 Opinion Retrieval from Blogs Wei Zhang, Clement Yu, and Weiyi Meng (2007 CIKM)
Week 10 Presentation Wesna LaLanne - REU Student Mahdi M. Kalayeh - Mentor.
WEEK 1 You have 10 seconds to name…
Amanda Lambert Jimmy Bobowski Shi Hui Lim Mentors: Brent Castle, Huijun Wang.
26/01/20161Gianluca Demartini Ranking Categories for Faceted Search Gianluca Demartini L3S Research Seminars Hannover, 09 June 2006.
SUPERBOWL ADVERTISING A Case Study in Buzz Marketing.
Slogans that Stick Just Do it- Nike This Bud is for you- Budweiser Have it your way- Burger King We bring good things to light- GE We’ll leave the light.
Learning Deep Rhetorical Structure for Extractive Speech Summarization ICASSP2010 Justin Jian Zhang and Pascale Fung HKUST Speaker: Hsiao-Tsung Hung.
Commercialization of IP in Suriname The current situation from a business perspective.
BUSINESS 1 Developing Products. BUSINESS 2 What Is a Product?  Features are the qualities, tangible and intangible, that a company builds into its products.
Automatic Advertisement Ratings Discussion Methods Problem and Motivation The goal is to automatically generate an objective score or ranking for an advertisement.
ETH 557 Week 2 DQ 2 To purchase this material link 557/ETH-557-Week-2-DQ-2 For more courses visit our website
Automatic Advertisement Rating
Example experiment 1: Which mobile plan will you sign up to?
Predicting Visual Search Targets via Eye Tracking Data
Textual Video Prediction Week 2
Query Based Video Summarization
Project #5: Generating Privacy and Security Threat Summary for Internet of Things REU Student: Ray Yan Graduate mentors: Logan Lebanoff Faculty Mentor(s):
Developing Your Personal Brand
Predicting Search Targets -Week 6
Done this week Image emotion classifier
REU Report Meetings – Week 7
McDonald’s Group 4 ).
Project 10: Innovative IoT Applications in Scientific Research and Consumer Market REU Student: Nathalie Domingo Graduate mentors: Kelvin Ly and Kaveh.
ECO 561 MART Perfect Education/ eco561mart.com.
ECO 561 OUTLET Perfect Education/ eco561outlet.com.
احد حیدری پژوهشگر و مشاور مهارتهای تجاری
Textual Video Prediction
Multiple Organ Detection in CT Volumes using CNN Week 4
Project # IoT Device Vulnerabilities and Security REU student: Amon Harris Graduate mentors: Orlando Arias Faculty mentor(s): Yier Jin, Shaojie Zhang.
Project 4: Internet of Things for Health and Wellness REU Student: Shanzila Chowdhury Graduate Mentor: Michael Eakin Faculty Mentor: Dr. David Metcalf.
Principal Component Analysis
Project 4: Internet of Things for Health and Wellness REU Student: Shanzila Chowdhury Graduate mentor: Michael Eakin Faculty Mentor: Dr. David Metcalf.
Project # Energy-Efficient Computing Devices for IoT REU student: Joseph Tobon Graduate mentors: Juan Escobedo Contreras and Stephen Williams Faculty.
Mentor: Salman Khokhar
Project 2: IoT Device Vulnerabilities and Security REU Student: Bryan Pearson Graduate mentors: Kelvin Ly and Kaveh Shamsi Faculty Mentor(s): Dr. Jin.
McDonald’s sell burgers, right?
Exploitations of Wireless Interfaces via Network Scanning
Project # IoT Device Vulnerabilities and Security REU student: Amon Harris Graduate mentors: Orlando Arias Faculty mentor(s): Yier Jin, Shaojie Zhang.
Project 10: Innovative IoT Applications in Scientific Research and Consumer Market REU Student: Nathalie Domingo Graduate mentors: Kelvin Ly and Kaveh.
Week 8 (July 1 – July 6, 2018) Accomplishments:
Байгууллагын хөгжил, соёл Г.Ганчимэг, PhD, MPH
Textual Video Prediction
Project # Energy-Efficient Computing Devices for IoT REU student: Joseph Tobon Graduate mentors: Juan Escobedo Contreras and Stephen Williams Faculty.
Warm Up Consumers of diet food products are concerned about the number of calories in each meal. A manufacturer advertises the number of calories in a.
Textual Video Prediction
Color-Attributes-Related Image Retrieval
Types of Errors And Error Analysis.
Project 6: Processing data from a visual IoT sensor of fluid/structure interactions with machine learning REU Students: Pete Orkweha & Alexis Downing.
REU: Week 7 TRECVID Sean McMillan.
Measurement Basics Chapter 13.
Deep screen image crop and enhance
… 1 2 n A B V W C X 1 2 … n A … V … W … C … A X feature 1 feature 2
Deep screen image crop and enhance
Slogan Analysis Senior B Mandy
Deep screen image crop and enhance
Deep screen image crop and enhance
REU Week 7: Real-Time Video Anomaly Detection
Presentation transcript:

Automatic Advertisement Rating Week 9 Mentors: Dr. Mubarak Shah, Dong Zhang, Amir Mazaheri REU Student: Brian Mora

This week Feature Extraction More Ranking SVM Experiments More SVR Experiments

Feature Extraction TRECVID Concepts 1x60 SentiBank Scores 1x1200

Ranking SVM Separate SVM for each brand (using likes/views for ranking) Single SVM for all brands (using likes/views for ranking)

Ranking SVM Results Separate SVM for each Brand (Reporting Error) Geico Heineken Mazda McDonald T-mobile Toyota Burger King Budweiser Coca-Cola Average Original Attributes 0.492963 0.492222 0.49848 0.500282 0.496405 0.502014 0.5050351 0.490476 0.49478 0.496961847 Refined Attributes 0.484667 0.530333 0.501193 0.500714 0.503548 0.505185 0.505185185 0.496264 0.494222 0.502367928 VGG 0.4 0.368421 0.380952 0.491228 0.602339 0.424836601 0.32967 0.421938624 TRECVID 0.3 0.7 0.431579 0.326316 0.614035 0.467836 0.503267974 0.395604 0.494505 0.470349327 E on random

Single SVM for All Brands Ranking SVM Results Single SVM for All Brands Original Attributes New Attributes VGG TRECVID Random Error 0.451307 0.531481 0.47952 0.521385 0.498035

SVR Results (Reporting RMSE) SVR (All Brands) Original Attributes Refined Attributes VGG TRECVID 0.00322 0.099973 0.004442 0.006017 Geico Heineken Mazda McDonald T-mobile Toyota Burger King Budweiser Coca-Cola Average Original Attributes 0.00204858 0.00203468688 0.00904026 0.00641063367 0.0077391895 0.00644922 0.045220523617 0.003924956966 0.00576923 0.009849 Refined Attributes 0.25393841 0.00242572937 0.23220679 0.09995315831 0.2913306084 0.21935099 0.192223929777 0.290131622539 0.29326522 0.208314 VGG 0.00190658 0.00308240648 0.00183597 0.00333060081 0.0024830461 0.00198366 0.007912630403 0.003636762882 0.00623102 0.0036 TRECVID 0.00268624 0.00302790663 0.00357497 0.00276562927 0.0027656292 0.00163338 0.014047658156 0.002534558858 0.00370088 0.004082

SVR Results