The World's Leading Provider of OEM Categorization Solutions
Database & Real-time Technology
RuleSpace offers a unique mix of technology that allows for the querying of databases of millions of pre-categorized web sites and an optional real-time content analysis engine. These two components work in conjunction to provide the industry’s standard of performance, accuracy and coverage.
Database Lookup
Web Categorization Database (WCD)
The RuleSpace Web Categorization Database (WCD) is the primary repository of years of URL discovery and categorization performed by RuleSpace. It is the largest, most accurate collection of website category information available on the market.
The RuleSpace WCD has millions of entries which are categorized at different “nodes” – host, domain, directory or page – based on our determination of the best way to rate the content. The vast majority (approximately 97%) are categorized at the “domain” level.
Each URL is assigned one or more of over 80 categories, including 12 categories specific to sites optimized for viewing on wireless/mobile devices. RuleSpace currently provides coverage in 17 languages, and adds new languages every year.
The WCD is maintained by RuleSpace via a daily incremental update process. Millions of new sites are added each year, and the database is kept clear of dead sites through periodic purges —approximately 5 to 7 million sites are removed annually.
Additional Databases
RuleSpace offers a number of optional databases to supplement the primary WCD. These include:
- The Internet Watch Foundation’s (IWF) Child Abuse Image Content (CAIC) database of sites known to host illegal child abuse content
- Known phishing sites database
- Known malware sites database
Real-time Categorization
RuleSpace Web Categorization Real-time Engine (WCRE)
The RuleSpace Web Categorization Real-time Engine (WCRE) is an optional supplement to the database lookup. It is designed to offer an additional level of analysis for unknown sites (i.e. sites not in the WCD) as well as sites with user generated dynamic content like search engines, social networking or blogging sites.
The WCRE examines a web site for content in over 30 specific category/language pairs (e.g. “Pornography, German” or “Anonymizer, English”) and if found, will return the appropriate category.
Additional Real-time Engines
RuleSpace offers a number of optional real-time engines. These include:
- Web Categorization Topic Detector to identify web content containing specific term(s) or phrases (e.g. obscene language).
- Image Categorization to analyse and identify pornographic images
