Annotation for free: Video tagging by mining user search behavior
The problem of tagging is mostly considered from the perspectives of machine learning and data-driven philosophy. A fundamental issue that underlies the success of these approaches is the visual similarity, ranging from the nearest neighbor search to manifold learning, to identify similar instances...
Saved in:
Main Authors: | , , , |
---|---|
Format: | text |
Language: | English |
Published: |
Institutional Knowledge at Singapore Management University
2013
|
Subjects: | |
Online Access: | https://ink.library.smu.edu.sg/sis_research/6467 https://ink.library.smu.edu.sg/context/sis_research/article/7470/viewcontent/2502081.2502085.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-7470 |
---|---|
record_format |
dspace |
spelling |
sg-smu-ink.sis_research-74702022-01-10T06:01:43Z Annotation for free: Video tagging by mining user search behavior TING, Yao MEI, Tao NGO, Chong-wah LI, Shipeng The problem of tagging is mostly considered from the perspectives of machine learning and data-driven philosophy. A fundamental issue that underlies the success of these approaches is the visual similarity, ranging from the nearest neighbor search to manifold learning, to identify similar instances of an example for tag completion. The need to searching for millions of visual examples in high-dimensional feature space, however, makes the task computationally expensive. Moreover, the results can suffer from robustness problem, when the underlying data, such as online videos, are rich of semantics and the similarity is difficult to be learnt from low-level features. This paper studies the exploration of user searching behavior through click-through data, which is largely available and freely accessible by search engines, for learning video relationship and applying the relationship for economic way of annotating online videos. We demonstrated that, by a simple approach using co-click statistics, promising results were obtained in contrast to feature-based similarity measurement. Furthermore, considering the long tail effect that few videos dominate most clicks, a new method based on polynomial semantic indexing is proposed to learn a latent space for alleviating the sparsity problem of click-through data. The proposed approaches are then applied for three major tasks in tagging: tag assignment, ranking, and enrichment. On a bipartite graph constructed from click-through data with over 15 million queries and 20 million video URL clicks, we showed that annotation can be performed for free with competitive performance and minimum computing resource, representing a new and promising paradigm for video tagging in addition to machine learning and data-driven methodologies. 2013-10-01T07:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/6467 info:doi/10.1145/2502081.2502085 https://ink.library.smu.edu.sg/context/sis_research/article/7470/viewcontent/2502081.2502085.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 Annotation Click-through data Tag assignment Tag enrich- ment Tag ranking Video search Video tagging Data Storage Systems Graphics and Human Computer Interfaces |
institution |
Singapore Management University |
building |
SMU Libraries |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
SMU Libraries |
collection |
InK@SMU |
language |
English |
topic |
Annotation Click-through data Tag assignment Tag enrich- ment Tag ranking Video search Video tagging Data Storage Systems Graphics and Human Computer Interfaces |
spellingShingle |
Annotation Click-through data Tag assignment Tag enrich- ment Tag ranking Video search Video tagging Data Storage Systems Graphics and Human Computer Interfaces TING, Yao MEI, Tao NGO, Chong-wah LI, Shipeng Annotation for free: Video tagging by mining user search behavior |
description |
The problem of tagging is mostly considered from the perspectives of machine learning and data-driven philosophy. A fundamental issue that underlies the success of these approaches is the visual similarity, ranging from the nearest neighbor search to manifold learning, to identify similar instances of an example for tag completion. The need to searching for millions of visual examples in high-dimensional feature space, however, makes the task computationally expensive. Moreover, the results can suffer from robustness problem, when the underlying data, such as online videos, are rich of semantics and the similarity is difficult to be learnt from low-level features. This paper studies the exploration of user searching behavior through click-through data, which is largely available and freely accessible by search engines, for learning video relationship and applying the relationship for economic way of annotating online videos. We demonstrated that, by a simple approach using co-click statistics, promising results were obtained in contrast to feature-based similarity measurement. Furthermore, considering the long tail effect that few videos dominate most clicks, a new method based on polynomial semantic indexing is proposed to learn a latent space for alleviating the sparsity problem of click-through data. The proposed approaches are then applied for three major tasks in tagging: tag assignment, ranking, and enrichment. On a bipartite graph constructed from click-through data with over 15 million queries and 20 million video URL clicks, we showed that annotation can be performed for free with competitive performance and minimum computing resource, representing a new and promising paradigm for video tagging in addition to machine learning and data-driven methodologies. |
format |
text |
author |
TING, Yao MEI, Tao NGO, Chong-wah LI, Shipeng |
author_facet |
TING, Yao MEI, Tao NGO, Chong-wah LI, Shipeng |
author_sort |
TING, Yao |
title |
Annotation for free: Video tagging by mining user search behavior |
title_short |
Annotation for free: Video tagging by mining user search behavior |
title_full |
Annotation for free: Video tagging by mining user search behavior |
title_fullStr |
Annotation for free: Video tagging by mining user search behavior |
title_full_unstemmed |
Annotation for free: Video tagging by mining user search behavior |
title_sort |
annotation for free: video tagging by mining user search behavior |
publisher |
Institutional Knowledge at Singapore Management University |
publishDate |
2013 |
url |
https://ink.library.smu.edu.sg/sis_research/6467 https://ink.library.smu.edu.sg/context/sis_research/article/7470/viewcontent/2502081.2502085.pdf |
_version_ |
1770575968090456064 |