Location-Based Keyword Recommendation via Virtual Region Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional search engines fail to recommend keywords based on a user's location, limiting the relevance of search results in mobile search environments.
Innovation Solution
A location-based keyword recommending system that collects location information, sets virtual regions through clustering, and recommends keywords frequently used within those regions, using a keyword collecting unit, region setting unit, and keyword recommending unit.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional keyword recommendation systems are used, then keyword recommendations are provided, but the recommendations are not based on user location and thus less relevant to mobile search needs
Solution Approach 1:
The patent applies local quality by making keyword recommendations location-specific. The system divides the search space into location-based segments and provides different keyword recommendations based on the user's geographical position. This is achieved through the region setting unit that creates virtual regions and the keyword recommending unit that selects keywords specific to each region, ensuring that recommendations are locally relevant rather than universally applied.
Solution Approach 2:
The patent implements dynamics by making the keyword recommendation system adaptable to changing user locations. The system dynamically adjusts keyword recommendations based on real-time location information obtained from mobile terminals. As users move between different virtual regions, the system dynamically updates the recommended keywords to match their current location, making the system flexible and responsive to spatial changes.
2Measurement precision
If location information is collected and virtual regions are set through clustering, then location-based keyword recommendations are achieved, but system complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the geographical space into discrete virtual regions through clustering algorithms. Instead of handling continuous location data, the system segments the search space into manageable regions, each with its own keyword set. This segmentation simplifies the complexity by transforming a continuous problem into discrete, manageable units that can be processed independently.
Solution Approach 2:
The patent introduces virtual regions as an intermediary layer between raw location information and keyword recommendations. Rather than directly mapping location coordinates to keywords, the system uses virtual regions as a mediating structure. This intermediary simplifies the relationship between location and keywords by providing a structured, clustered representation that bridges the gap between continuous spatial data and discrete keyword sets.
Data Source
AI summary
According to exemplary embodiments of the invention, a location-based keyword recommending system and method are provided. The location-based keyword recommending system may include a keyword collecting unit to store location information regarding a location where a keyword is input, a region setting unit to set a virtual region by performing clustering of the location information with reference to the keyword, a region combining unit to combine virtual regions overlapping each other into one virtual region, and a keyword recommending unit to provide a location-based keyword based on the keyword related to the location information of the virtual region.