Numeric web search

Conventional language based web search can be combined with numeric web search (vectorial similarity search). An as simple as possible searchable "pattern" structure is proposed, which consists of a "pattern name" and a feature vector. The pattern name identifies the meaning of the feature vector, which can precisely represent all measurable objects (and more). Similarity search is always done only over patterns with the same pattern name, because these are directly comparable. Similarities of patterns are mapped to spatial similarities (small distances) of their feature vectors using an appropriate metric. This makes them searchable. Pattern names can be "owned" like today domain names. This facilitates unbureaucratic definition of patterns: Every owner of an internet domain name dn can define all patterns with names dn/*. This would allow to mobilize a great working potential.
The pattern structure can be used to make all measurable objects (and more) searchable. One application would be medicine: If there is a new patient with certain diagnostic patterns, it would be possible to search in a global database for completed histories of patients with similar diagnostic patterns to find the best treatment.
Details are described in http://www.orthuber.com/wpa.htm

Category:

Science & Technology

Tags:

• numeric
• web
• search
• similarity
• metric
• databases
• feature
• vectors
• patterns
• pattern
• names
• medicine