A Relational Aggregated Disjoint Multimedia Search Results Approach using Semantics

The web holds a vast collection of diverse multimedia information. Web search engines are a major gateway to access these resources. Multimedia objects, including images, video, and audio, are increasing on the world wide web. This increasing content heterogeneity is transforming user information needs from factual searches to boarder multimedia information exploration. At present, pertinent multimedia information is scattered in various places. Traditional search engines bridge the content heterogeneity via aggregation techniques. Yet, this aggregation is mostly partial or non-blended, which causes a loss of retrieved multimedia contents inter-relationships. We proposed an relational aggregation search approach for the disjoint multimedia search results using semantics. Our relational aggregation approach assembles the disjoint multimedia vertical snippets, fully blend, and re-rank them in a linearly ranked list based on their semantic similarity with respect to the user's query. This helps retain the inter-relationship between multimedia contents and enhance multimedia searches.