Building a custom distance function for typographical errors - particularly with the QWERTY keyboard
In this mini project, I will extend/move away from the regular approaches to distance function, auto-susggest and spell checker feature already available in Python, thought I will periodically extrend the base class for graph traversals of neighbouring nodes, as in the keyboard. I treat the QWERTY keyboard as a graph with connected adjacent nodes,and using the common typos, lay out a structure between the 3 rows and connect them with appropriate distances. Post this I will implement the custom distance metric - which then I intend to compare for performance with other methods [Apache Solr, Spell checker & Auto-suggest (but only in context of similarity)]
For ideas, suggestions, collaboration, reach out to me at ekta1007@gmail.com