Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Shotgun HuDist Key Generation with Recursive Typeahead Lookup

IP.com Disclosure Number: IPCOM000218273D
Publication Date: 2012-May-31
Document File: 6 page(s) / 84K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a method applied to HuDist key generation which provides users with a completely distinct typeahead lookup in user-friendly, typically typed terms derived from and resolvable to individual records in a database for which the user is searching. The disclosed solution uses recursive HuDist, taking user input and providing distinct typeahead values using the fewest number of terms to identify one and only one record.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 38% of the total text.

Page 01 of 6

Shotgun HuDist Key Generation with Recursive Typeahead Lookup

A HuDist scheme is an aggregation of result sets (based on HuDist Keys), typically correlated with a foreign key in a relational database , which allows a user to uniquely distinguish amongst records. Where a foreign key in the database might present a user with a selection list of actual foreign key values (e.g., 10401, 041320, 04410), a HuDist scheme presents the user with recognizable terms (e.g., Ted Johnson, Terry R. McIntosh, Terry Z. McIntosh), mapped to the actual foreign key values. In plain language, a foreign key can be thought of as a constrained list from which a user will select - a list constrained by what is actually in the underlying database . A HuDist Scheme addresses two specific needs: providing users with values they can understand, and providing a selection list derived from the actual database (rather than from previous 'popular' search terms) which is completely distinct (technically unique) and distinguishable (by a human).

The term "scheme" in this context refers to a set of one or more HuDist Keys, as well as the Foreign Key which it supplants, in the context of user selection. The HuDist is special due to the use of multiple keys in an iterative algorithm dynamically controlled by context and content.

The problem is in providing users with a completely distinct typeahead lookup in user-friendly, typically typed terms (which would be resolved to one, and only one, surrogate key value) against very large databases in which duplication exists with varying intensity, dependent on context (e.g,. in the subset of Chinese people in a database of people, the surname "Li" may be very common and "Smith" may be unique, while in the English subset, "Smith" may be common and "Li" may be unique).

A HuDist Scheme provides human-understandable terms. "Shotgunning" the HuDist, whereby the HuDist is broken up into fragments and reassembled in every possible combination in every possible order , provides a means to support every possible way a user might have of entering the value. The number of possibilities, given a set of elements in a shotgun set, can be computed as follows:
n_P_n! (Based on n_P_k, number of permutations in k given n) so for a set of 3 elements n, the possibilities are :

3_P_3! =

(3!/(3-3)!) + (3!/(3-2)!) + (3!/(3-1)!) = 15

For example, a HuDist key might be created for a People table using First Name (fn), Last Name (ln), City. This key can be shotgunned to produce these fifteen HuDist keys:

fn
fn, ln
fn, ln, city

1


Page 02 of 6

fn, city, ln fn, city
ln
ln, fn
ln, fn, city ln, city, fn ln, city city, fn, ln city, ln, fn city, fn city, ln city

The disclosed solution uses recursive HuDist , taking user input and providing distinct typeahead values using the fewest number of terms.

In the example above, if a user types "Amy L", and the typeahead dynamically renders a list of "Amy L"s, the Recursive HuDist ensures that, for two people both named A...