Geocoding Data Sanitization Using Normalized Levenshtein Distance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Geocoding processes are often inaccurate due to incomplete or poorly formatted geolocation data from disparate databases, leading to incorrect plotting of points of interest on maps, with 25% of hotels being incorrectly located over a mile away from their actual position.
Innovation Solution
The technology employs data sanitization and normalization methods, including converting segments into standardized formats, calculating a normalized Levenshtein distance, and comparing distance scores to thresholds to identify and correct discrepancies in geodata from different sources, ensuring accurate plotting of points of interest.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If geocoding processes use data from disparate databases with different formats, then data coverage and quantity increase, but data quality and accuracy deteriorate
Solution Approach 1:
The patent segments geocoding into distinct phases: data collection from multiple databases, format normalization, quality validation, and error correction. Each phase handles specific aspects of data processing independently, allowing comprehensive data coverage while maintaining accuracy through specialized processing at each stage
Solution Approach 2:
The system changes data parameters by applying format normalization rules that convert diverse database formats into a standardized structure. This includes transforming address formats, coordinate systems, and data types across different sources while preserving the underlying geographic information accuracy
2Quantity of substance
If data from multiple sources is integrated without normalization, then data quantity increases, but data consistency deteriorates
Solution Approach 1:
The patent implements a universal normalization framework that processes data from multiple database sources through common format standards. This multi-functional system handles various data types (addresses, coordinates, point of interest data) from different sources using unified transformation rules, ensuring consistent output regardless of input source
Solution Approach 2:
The system applies homogeneity by enforcing standardized data formats across all integrated sources. All geocoded data undergoes transformation to match a consistent structure including uniform address formatting, coordinate representations, and metadata schemas, eliminating format-related inconsistencies
3Speed
If traditional geocoding methods are used, then processing speed is maintained, but geocoding accuracy deteriorates
Solution Approach 1:
The patent applies preliminary actions by pre-processing data through format normalization and validation before the actual geocoding operation. Common formatting issues are corrected in advance, and data quality checks are performed beforehand, enabling faster and more accurate geocoding without requiring complex post-processing corrections
Solution Approach 2:
The system introduces an intermediary normalization layer between raw data ingestion and geocoding processing. This intermediate stage standardizes data formats and validates quality metrics, acting as a mediator that prepares data for accurate geocoding while maintaining processing efficiency through automated transformation rules
Data Source
AI summary
Embodiments of the present technology relate to data sanitization and normalization and geocoding methods that apply the same. An example method includes sanitizing geodata sets and normalizing the sanitized geodata using a normalized Levenshtein distance Algorithm.