Faceted topic retrieval of news video using joint topic modeling of visual features and speech transcripts
Because of the inherent ambiguity in user queries, an important task of modern retrieval systems is faceted topic retrieval (FTR), which relates to the goal of returning diverse or novel information elucidating the wide range of topics or facets of the query need. We introduce a generative model for...
Saved in:
Main Authors: | , , , |
---|---|
Format: | text |
Language: | English |
Published: |
Institutional Knowledge at Singapore Management University
2010
|
Subjects: | |
Online Access: | https://ink.library.smu.edu.sg/sis_research/6875 https://ink.library.smu.edu.sg/context/sis_research/article/7878/viewcontent/Faceted_topic_retrieval_of_news_video_using_joint_topic_modeling_of_visual_features_and_speech_transcripts.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Singapore Management University |
Language: | English |
id |
sg-smu-ink.sis_research-7878 |
---|---|
record_format |
dspace |
spelling |
sg-smu-ink.sis_research-78782022-02-07T11:07:03Z Faceted topic retrieval of news video using joint topic modeling of visual features and speech transcripts WAN, Kong-Wah TAN, Ah-hwee LIM, Joo-Hwee CHIA, Liang-Tien Because of the inherent ambiguity in user queries, an important task of modern retrieval systems is faceted topic retrieval (FTR), which relates to the goal of returning diverse or novel information elucidating the wide range of topics or facets of the query need. We introduce a generative model for hypothesizing facets in the (news) video domain by combining the complementary information in the visual keyframes and the speech transcripts. We evaluate the efficacy of our multimodal model on the standard TRECVID-2005 video corpus annotated with facets. We find that: (1) the joint modeling of the visual and text (speech transcripts) information can achieve significant F-score improvement over a text-alone system; (2) our model compares favorably with standard diverse ranking algorithms such as the MMR [1]. Our FTR model has been implemented on a news search prototype that is undergoing commercial trial. 2010-07-01T07:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/6875 info:doi/10.1109/ICME.2010.5583061 https://ink.library.smu.edu.sg/context/sis_research/article/7878/viewcontent/Faceted_topic_retrieval_of_news_video_using_joint_topic_modeling_of_visual_features_and_speech_transcripts.pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University faceted topic retrieval multimedia topic modeling latent Dirichlet allocation Artificial Intelligence and Robotics Databases and Information Systems |
institution |
Singapore Management University |
building |
SMU Libraries |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
SMU Libraries |
collection |
InK@SMU |
language |
English |
topic |
faceted topic retrieval multimedia topic modeling latent Dirichlet allocation Artificial Intelligence and Robotics Databases and Information Systems |
spellingShingle |
faceted topic retrieval multimedia topic modeling latent Dirichlet allocation Artificial Intelligence and Robotics Databases and Information Systems WAN, Kong-Wah TAN, Ah-hwee LIM, Joo-Hwee CHIA, Liang-Tien Faceted topic retrieval of news video using joint topic modeling of visual features and speech transcripts |
description |
Because of the inherent ambiguity in user queries, an important task of modern retrieval systems is faceted topic retrieval (FTR), which relates to the goal of returning diverse or novel information elucidating the wide range of topics or facets of the query need. We introduce a generative model for hypothesizing facets in the (news) video domain by combining the complementary information in the visual keyframes and the speech transcripts. We evaluate the efficacy of our multimodal model on the standard TRECVID-2005 video corpus annotated with facets. We find that: (1) the joint modeling of the visual and text (speech transcripts) information can achieve significant F-score improvement over a text-alone system; (2) our model compares favorably with standard diverse ranking algorithms such as the MMR [1]. Our FTR model has been implemented on a news search prototype that is undergoing commercial trial. |
format |
text |
author |
WAN, Kong-Wah TAN, Ah-hwee LIM, Joo-Hwee CHIA, Liang-Tien |
author_facet |
WAN, Kong-Wah TAN, Ah-hwee LIM, Joo-Hwee CHIA, Liang-Tien |
author_sort |
WAN, Kong-Wah |
title |
Faceted topic retrieval of news video using joint topic modeling of visual features and speech transcripts |
title_short |
Faceted topic retrieval of news video using joint topic modeling of visual features and speech transcripts |
title_full |
Faceted topic retrieval of news video using joint topic modeling of visual features and speech transcripts |
title_fullStr |
Faceted topic retrieval of news video using joint topic modeling of visual features and speech transcripts |
title_full_unstemmed |
Faceted topic retrieval of news video using joint topic modeling of visual features and speech transcripts |
title_sort |
faceted topic retrieval of news video using joint topic modeling of visual features and speech transcripts |
publisher |
Institutional Knowledge at Singapore Management University |
publishDate |
2010 |
url |
https://ink.library.smu.edu.sg/sis_research/6875 https://ink.library.smu.edu.sg/context/sis_research/article/7878/viewcontent/Faceted_topic_retrieval_of_news_video_using_joint_topic_modeling_of_visual_features_and_speech_transcripts.pdf |
_version_ |
1770576111302868992 |