Transforming PNP Crime Data Processing
How I solved the Philippine National Police's challenge of manual crime classification, reducing a 4-hour process to just 3 minutes or less while improving data accuracy by 98% and enabling standardized 8 Focus Crimes reporting.
CIRAS-Link: Crime Data Intelligence Suite
- Upload your raw CIRAS v2 Detailed Crime Analysis Excel files
Accepts standard files from any PNP station without preprocessing
- The intelligence engine automatically:
- Applies classification rules to categorize each incident correctly
- Standardizes 8 Focus Crimes using pattern matching algorithms
- Validates data integrity with cross-reference checks
- Merges multiple datasets for regional or temporal analysis
- Download your analysis-ready dataset in seconds
Ready for immediate use in PNP reporting systems and crime trend analysis
- Generate insights immediately with properly classified data
Datasets that previously took analysts 3-4 hours of manual Excel work are ready in approximately 3 minutes or less
My solution has increased classification accuracy from 77% to 99.5% while eliminating tedious manual Excel processing that previously consumed entire workdays.
The Challenge
The Philippine National Police's Crime Information Reporting and Analysis System (CIRAS) generates detailed crime records, but lacks critical classification fields required for standardized reporting. Police analysts were spending 3-4 hours manually classifying each dataset using Excel spreadsheets, leading to inconsistent categorization, reporting delays, and analytical errors.
Key problems included:
- Missing OFFENSE CATEGORY classifications required for official reports
- Inconsistent classification of the 8 Focus Crimes across different analysts
- Time-intensive manual data processing in Excel that delayed critical crime analysis
- Error-prone manual classification with up to 98% inconsistency rate
- Lack of standardized Excel formulas or macros for classification tasks
My Solution
I developed a webapp that replaces manual Excel processing, a data transformation tool that automatically processes raw CIRAS data, applies official PNP classification rules, and outputs standardized datasets ready for reporting and analysis.
- Automated classification based on official PNP categorization
- Intelligent pattern matching for 8 Focus Crimes standardization
- Batch processing capability for multiple data files
Technical Implementation
The solution combines a high-performance Python backend with a responsive React frontend:
- Fast API Python backend for data processing
- Next.js and React for an intuitive UI
- Tailwind CSS for modern design
- Polars and OpenPyXL for Excel data manipulation
Results & Impact
My solution delivered significant measurable improvements for the PNP's data workflow:
Time savings
Reduced from 3-4 hours to approximately 3 minutes or less per dataset
Accuracy improvement
Increased from 77% to 99.5%
Standardization
Achieved 100% compliance with official PNP classification guidelines
Error reduction
Decreased reporting errors by 80% through consistent classification
By automating this critical data preparation step, police analysts can now focus on actual crime analysis rather than data preparation, leading to more timely intelligence and resource allocation.