blue banner
//

Graduate Projects - Details

Computer Science Program

Project ID: 483
Author: Sumanth Kandakatla
Project Title: AN APPLICATION FOR FOOD PHOTO MINING USING GEO-PHOTO TWEETS
Semester: 1 2015
Committe Chair: Dr. Longzhuang Li
Committee Member 1: Dr. David Thomas
Project Description: Micro-blogging is currently a major trend. Most of the users normally write down their daily experience, feeling or emotion on their wall. This includes uploading photos of food they have taken. So many users are posting photos as well as short messages to Twitter every few minutes from everywhere on the earth. By observing the Twitter stream, we can achieve various types of photos with texts. In this proposed system, we develop a web application where the admin will connect to Twitter and collect the tweets and analyse the tweets, storing the tweets and displaying the photos on google maps. For analysing the tweets, apache Lucene [15] is used for sentiment analysis of tweets that are collected. And for collecting food photos from Twitter we monitor Twitter stream having both food-related keywords and photos also called as geo-photo tweets, then apply a foodness classifier. In our proposed system I use support vector machine classifier [14], [17] to classify whether the obtained photos are food photos or not. In addition to that, this system will allows users to search a particular food and produce the nearby restaurant on google map just by clicking on the food photo that has been obtained through the collected tweets. And this web application allows users to post feedback on the food for a restaurant that are provided from google maps.
Project URL:   483.pdf
 
© Texas A&M University-Corpus Christi • 6300 Ocean Drive, Corpus Christi, Texas 78412 • 361-825-5700