IP Library › Granted Patent US 10,102,272
Granted Patent B2
US 10,102,272 · App. 15/208,593 · Granted Oct 16, 2018

System and method for ranking documents

Inventor: Aravind Musuluri (Birmingham, AL)
G06F17/30616
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,102,272
App. No.
15/208,593
Granted
Oct 16, 2018
Kind
B2
Abstract

Systems and methods for enhancing user experience in a search environment are provided. The method includes ranking documents which have multiple zones wherein one zone in a document dominates rest of the zones in the document.

Claims (16)

1. A computer implemented method of ranking results of a search operation comprising:

(a) accepting a search query comprising search query terms;

(b) identifying documents comprising the search query terms;

(c) for each document, computing a search query term score for each search query term, wherein the search query term score is a dominant zone score or a non-dominant zone score for a matching term in the document calculated by; (c1) dividing said document into dominant and non-dominant zones; (c2) calculating number of distinct terms in the dominant zone and assigning a dominant zone score to the distinct terms, such that the dominant zone term score is inversely proportional to the number of distinct terms in said zone; (c3) assigning a non-dominant zone score for distinct terms in the non-dominant zones that are not present in the dominant zones; and (c4) computing a search query term score for each search query term, wherein the search query term score is a dominant zone score or a non-dominant score for a matching term in document;

(d) computing a document score for each document wherein the document score is a sum total of the matched search query term scores; and

(e) ranking the documents based on the computed document score for each document.

2. The method as in claim 1 , wherein the dominant zone term scores and the non-dominant zone term scores are pre-computed and stored in an index.

3. The method as in claim 1 , wherein the zones are tokenized to terms prior to assigning a dominant zone score or a non-dominant zone score to each term.

4. The method as in claim 3 , wherein the tokenization comprises removing less value adding words.

5. The method as in claim 1 , wherein the dominant zone is processed said method comprising identifying adjectives in the dominant zone and treating the adjectives as if the adjectives are part of non-dominant zone.

6. The method as in claim 1 , wherein the documents with a higher document score are ranked higher.

7. The method as in claim 1 , where in the documents with the same document score may be further ranked based on frequency and/or popularity of non-query terms in the dominant zone.

8. The method as in claim 1 , wherein the dominant zone is a title.

9. The method as in claim 1 , wherein the search query is tokenized into terms.

10. The method as in claim 1 , wherein the dominant zones of said document are combined into one dominant zone.

11. The method as in claim 1 , wherein the non-dominant zone of said document are combined into one non-dominant zone.

Continuity (2)
Provisional Application 62191486 · Jul 12, 2015
Related Publication 20170011043A1 · Jan 12, 2017
Cited By (1)
US 12,566,804