Please use this identifier to cite or link to this item: http://hdl.handle.net/123456789/29118
Title: Aggregated Vertical Web Search Results Research Framework
Authors: Sumaira Ambreen
Keywords: Computer Sciences
Issue Date: 2023
Publisher: Quaid I Azam University Islamabad
Abstract: Conventional search engines often present users with ranked lists of search results, necessitating manual sifting through documents to extract infor mation. This approach restricts the exploration of web-based multimedia content, particularly when using vertical search engines. Interacting with results from various verticals can lead to a loss of exploration context, mak ing it challenging to assemble relevant information and requiring extensive scrolling and clicking. To address these limitations, we propose a concise framework for research ing of aggregated vertical web search results, facilitating comprehensive exploration of aggregated multimedia documents. Our system employs ad vanced techniques such as clustering and summarization to efficiently or ganize search results and enhance user interaction. Non-linear representa tions, such as tree-map and pie-chart visualizations, offer an intuitive and interactive exploration experience. The proposed system’s user-friendly in terface enables seamless exploration across disjoint verticals, maximizing contextual understanding and cognitive engagement. A comprehensive assessment of the proposed aggregated research frame work, tool, and mechanism has been undertaken for the purpose of gen eral usability research and empirical analysis. The collective findings vividly demonstrate the precision and feasibility of the aggregated research frame iv work. The user interface (SUI) of the proposed tool proves to be user friendly, delivering a high level of satisfaction while minimizing the number of clicks required and seconds spent
URI: http://hdl.handle.net/123456789/29118
Appears in Collections:M.Phil

Files in This Item:
File Description SizeFormat 
COM 2533.pdfCOM 25331.3 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.