Efficient processing of top-k frequent spatial keyword queries.

Area of Science:

• Data Science
• Geographic Information Systems
• Database Management

Background:

• High-speed mobile communication and network devices generate spatial textual big data (STBD) with complex attributes.
• Existing top-k spatial keyword query methods primarily index location and few attributes, neglecting inter-attribute correlations.
• Data mining from STBD is crucial for extracting valuable insights from diverse spatial information.

Purpose of the Study:

• To address the gap in analyzing interrelationships within textual attributes of STBD.
• To define and develop algorithms for the top-k frequent spatial keyword query (tfSKQ).
• To propose an efficient indexing structure for handling complex spatial multi-keyword queries.

Main Methods:

• Defined the top-k frequent spatial keyword query (tfSKQ).
• Proposed a novel hybrid index structure, the RCL-tree, grounded in concept lattice theory.
• Developed tfSKQ algorithms for retrieving frequent and nearest spatial objects.

Main Results:

• Experimental evaluation using real datasets demonstrated the effectiveness of the proposed approach.
• The RCL-tree significantly improved performance in tfSKQ under complex query conditions.
• The study successfully retrieved the most frequent and nearest spatial objects from STBD.

Conclusions:

• The proposed RCL-tree and tfSKQ algorithms offer an effective and efficient solution for spatial textual big data analysis.
• This research enhances the capability to explore interrelated knowledge within textual attributes of spatial data.
• The findings contribute to advancing data mining techniques for complex spatial multi-keyword queries.