Multimodal data fusion in text-image heterogeneous graph for social media recommendation

  • Yu Xiong
  • , Daling Wang
  • , Yifei Zhang
  • , Shi Feng
  • , Guoren Wang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

7 Citations (Scopus)

Abstract

Every day, millions of texts, images, audios, videos, and other information with different modalities are posted on social media. These multimodal data provide abundant resources for information recommendation. In this paper, a new method based on multimodal data fusion is proposed for more effective recommendation on social media. Firstly, a heterogeneous graph on texts and images is created effectively to represent the relationship of multimodal data. Then the relationship of multimodal data is fused based on graph clustering to improve the quality of social media recommendation. Finally, the multimodal social media information recommendation is performed as a process of walk on the proposed heterogeneous graph. The experiment on texts and images of microblogs shows social media recommendation using multimodal data fusion is better than that on single modality.

Original languageEnglish
Title of host publicationWeb-Age Information Management - 15th International Conference, WAIM 2014, Proceedings
PublisherSpringer Verlag
Pages96-99
Number of pages4
ISBN (Print)9783319080093
DOIs
Publication statusPublished - 2014
Externally publishedYes
Event15th International Conference on Web-Age Information Management, WAIM 2014 - Macau, China
Duration: 16 Jun 201418 Jun 2014

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume8485 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference15th International Conference on Web-Age Information Management, WAIM 2014
Country/TerritoryChina
CityMacau
Period16/06/1418/06/14

Keywords

  • data fusion
  • graph clustering
  • heterogeneous graph
  • multimodal data
  • recommendation

Fingerprint

Dive into the research topics of 'Multimodal data fusion in text-image heterogeneous graph for social media recommendation'. Together they form a unique fingerprint.

Cite this