Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

LDADeep+: Latent Aspect Discovery with Deep Representations

LDADeep+ utilizes the high-level meaning of deep learning representation, and combines it with topic model to learn good aspects
Abstract: 

Nowadays, with the success and fast growth of social media communities and mobile devices, people are encouraged to share their multimedia data online. Analyzing and summarizing data into useful information thus becomes increasingly important. For on- line photo sharing services like Flickr, when users are uploading a batch of daily photos at a time, the tags users provided tend to be rather vague, containing only a small amount of information. For better photo application and understanding, we attempt to automat- ically discover semantic-rich (hidden) aspects of photos merely by looking at image contents. In this paper, we propose an effective model, which is a combination of LDA model and deep learning rep- resentations, to realize the idea of automatic aspect discovery. We then discuss the properties of this aspect discovery model through experiments on event summarization task . In those experiments, we show the high diversity and high quality of aspects discovered by our proposed method. Meanwhile, we conduct an user study to evaluate the quality of the summarized results. Moreover, the pro- posed method can be further extended to human attribute discovery for a given event. We automatically discover different aspects on our Olympic Games data (e.g. football, ice skating).

up

Paper Details

Authors:
Chieh-En Tsai, Hui-Lan Hsieh, Winston Hsu
Submitted On:
24 March 2016 - 12:09pm
Short Link:
Type:
Presentation Slides
Event:
Presenter's Name:
Chieh-En Tsai (Andy Tsai)
Paper Code:
MMSP-L2.2
Document Year:
2016
Cite

Document Files

tsai_LDADeep_4_3 copy.pptx

(254 downloads)

Subscribe

[1] Chieh-En Tsai, Hui-Lan Hsieh, Winston Hsu, "LDADeep+: Latent Aspect Discovery with Deep Representations", IEEE SigPort, 2016. [Online]. Available: http://sigport.org/1028. Accessed: Dec. 17, 2017.
@article{1028-16,
url = {http://sigport.org/1028},
author = {Chieh-En Tsai; Hui-Lan Hsieh; Winston Hsu },
publisher = {IEEE SigPort},
title = {LDADeep+: Latent Aspect Discovery with Deep Representations},
year = {2016} }
TY - EJOUR
T1 - LDADeep+: Latent Aspect Discovery with Deep Representations
AU - Chieh-En Tsai; Hui-Lan Hsieh; Winston Hsu
PY - 2016
PB - IEEE SigPort
UR - http://sigport.org/1028
ER -
Chieh-En Tsai, Hui-Lan Hsieh, Winston Hsu. (2016). LDADeep+: Latent Aspect Discovery with Deep Representations. IEEE SigPort. http://sigport.org/1028
Chieh-En Tsai, Hui-Lan Hsieh, Winston Hsu, 2016. LDADeep+: Latent Aspect Discovery with Deep Representations. Available at: http://sigport.org/1028.
Chieh-En Tsai, Hui-Lan Hsieh, Winston Hsu. (2016). "LDADeep+: Latent Aspect Discovery with Deep Representations." Web.
1. Chieh-En Tsai, Hui-Lan Hsieh, Winston Hsu. LDADeep+: Latent Aspect Discovery with Deep Representations [Internet]. IEEE SigPort; 2016. Available from : http://sigport.org/1028