[Paper Review] Intelligent Agent Based Semantic Web in Cloud Computing Environment

Debajyoti Mukhopadhyay, Manoj Kumar Sharma+3 more|arXiv (Cornell University)|May 4, 2013

Semantic Web and Ontologies13 references3 citations

TL;DR

This paper proposes SemanTelli, a meta-semantic-search engine built on intelligent agents within a cloud computing environment to overcome the limitations of keyword-based search engines. By orchestrating multiple semantic search engines like Hakia, DuckDuckGo, and SenseBot through autonomous agents, SemanTelli enhances result relevance and efficiency through semantic understanding and distributed processing in the cloud.

ABSTRACT

Considering today's web scenario, there is a need of effective and meaningful search over the web which is provided by Semantic Web. Existing search engines are keyword based. They are vulnerable in answering intelligent queries from the user due to the dependence of their results on information available in web pages. While semantic search engines provides efficient and relevant results as the semantic web is an extension of the current web in which information is given well defined meaning. MetaCrawler is a search tool that uses several existing search engines and provides combined results by using their own page ranking algorithm. This paper proposes development of a meta-semantic-search engine called SemanTelli which works within cloud. SemanTelli fetches results from different semantic search engines such as Hakia, DuckDuckGo, SenseBot with the help of intelligent agents that eliminate the limitations of existing search engines.

Motivation & Objective

• To address the limitations of keyword-based search engines that rely on syntactic matching rather than semantic meaning.
• To enhance search relevance and efficiency by leveraging semantic web technologies in a distributed cloud environment.
• To design a meta-semantic-search engine that aggregates results from multiple existing semantic search engines.
• To implement intelligent agents that autonomously retrieve, rank, and combine results from diverse semantic sources.
• To demonstrate the feasibility and performance improvement of a cloud-hosted, agent-driven semantic search architecture.

Proposed method

• The system employs intelligent agents to dynamically query multiple external semantic search engines, including Hakia, DuckDuckGo, and SenseBot.
• Agents extract and normalize results using semantic metadata, enabling cross-engine comparison and fusion.
• A centralized ranking module applies a custom page-ranking algorithm to combine results based on relevance and source credibility.
• The architecture is deployed in a cloud environment to ensure scalability, fault tolerance, and on-demand resource allocation.
• Semantic annotations are used to enrich query understanding and improve result precision beyond keyword matching.
• The system uses a service-oriented design to enable modular integration of new semantic search engines and agents.

Experimental results

Research questions

• RQ1 How can intelligent agents be effectively used to aggregate results from multiple heterogeneous semantic search engines?
• RQ2 What architectural design enables efficient, scalable, and semantic-aware search in a cloud computing environment?
• RQ3 To what extent does combining results from multiple semantic engines improve search relevance compared to individual engines?
• RQ4 How do intelligent agents enhance the performance and adaptability of meta-search systems in dynamic web environments?
• RQ5 What role does cloud infrastructure play in supporting real-time, large-scale semantic search operations?

Key findings

• The proposed SemanTelli system successfully integrates results from multiple semantic search engines using intelligent agents, demonstrating improved result aggregation and relevance.
• The cloud-based deployment enables dynamic scaling and high availability, supporting real-time query processing.
• The use of semantic metadata and agent-based coordination reduces redundancy and increases precision in result fusion.
• The system's meta-ranking algorithm outperforms individual engine results by leveraging cross-source relevance signals.
• The architecture proves feasible for deployment in distributed, heterogeneous environments with minimal latency overhead.
• The integration of agents with cloud infrastructure enables autonomous, adaptive, and scalable semantic search operations.