Measuring semantic relatedness between Flickr images: from a social tag based view

Area of Science:

• Computer Vision
• Data Mining
• Information Retrieval

Background:

• Multimedia relatedness is crucial for applications like clustering, searching, recommendation, and annotation.
• Social media platforms generate vast amounts of user-tagged image data, necessitating efficient semantic analysis.

Purpose of the Study:

• To develop a robust method for measuring the semantic relatedness of images on platforms like Flickr.
• To improve the accuracy and efficiency of multimedia data analysis and retrieval.

Main Methods:

• Utilized four information theory-based functions to assess tag semantic relatedness.
• Employed bipartite graphs for integrating tag pairs, reducing noise and redundancy.
• Incorporated tag order information to emphasize high-position tags, enhancing semantic representation.

Main Results:

• Evaluated the proposed method on a dataset of 1000 Flickr images.
• Demonstrated effectiveness and robustness in data mining tasks such as image clustering and searching.
• Showcased successful applications in web-based tasks like faceted exploration.

Conclusions:

• The proposed method effectively measures semantic relatedness in Flickr images.
• The approach shows significant promise for various web-based multimedia applications.
• This work contributes to advancing semantic analysis in computer vision and data mining.