Page 1 |
Save page Remove page | Previous | 1 of 199 | Next |
|
small (250x250 max)
medium (500x500 max)
large ( > 500x500)
Full Resolution
All (PDF)
|
This page
All
Subset |
METHODS FOR IMPROVING SEARCH ON
U.S. PATENT DOCUMENTS
by
Shahzad Majeed Tiwana
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(COMPUTER SCIENCE)
May 2010
Copyright 2010 Shahzad Majeed Tiwana
Object Description
| Title | Methods for improving search on U.S. patent documents |
| Author | Tiwana, Shahzad Majeed |
| Author email | tiwana@usc.edu; shahzad.tiwana@gmail.com |
| Degree | Doctor of Philosophy |
| Document type | Dissertation |
| Degree program | Computer Science |
| School | Viterbi School of Engineering |
| Date defended/completed | 2009-11-05 |
| Date submitted | 2010 |
| Restricted until | Unrestricted |
| Date published | 2010-05-11 |
| Advisor (committee chair) | Horowitz, Ellis |
| Advisor (committee member) |
Boehm, Barry Lu, Stephen |
| Abstract | Patents are structured documents that contain important information related to a certain invention and purport to describe the invention in very precise terms. Patent search is often conducted by inventors, patent attorneys, technical and business experts to find the prior art and mitigate risks. Prior art searches are the most common searches and are performed before filing an application to ascertain patentability of an invention, during the application process to determine novelty of the invention, to invalidate a patent’s claim of originality or to learn about a specific field of invention. Due to the complex nature of the Patent document, traditional information retrieval approaches (IR) do not perform very well.; In this thesis, I investigate methods to improve patent document information retrieval. I propose a tunable and parametric citation based algorithm “FindCite”, that can be used for easily collecting relevant prior art patents from a patent dataset. Using citations, this algorithm can discover relevant patents even if they do not contain the key words or phrases issued as search queries. Our experiments demonstrate that FindCite results in better precision and recall as compared to other publicly available patent search systems including USPTO patent search and Google patent search.; Additionally, I use “Problem – Solution approach” for patent document information retrieval, in which I investigate methods to identify the problem that a certain invention solves. I propose new methods to automatically extract problem solved concepts, which is a statement of “What problem does the invention solve” from the patent documents. The proposed approaches require no training, thus are domain independent and can be used in any text based information extraction or information retrieval system.; Finally, I propose Patent Document Markup Language (PDML), a descriptive and referential markup which allows marking up and linkage of the problems and their solutions described in the patent documents. PDML not only facilitates navigation from a specific problem to its solution and the claims within a patent document, but also facilitates discovery of relevant patents across the patent dataset; for example, it makes it possible to find all patents that provide different solutions for a certain problem. |
| Keyword | aspect synonyms; information retrieval; patent; Patent Document Markup Language; prior art patents; search |
| Geographic subject (country) | USA |
| Language | English |
| Part of collection | University of Southern California dissertations and theses |
| Publisher (of the original version) | University of Southern California |
| Place of publication (of the original version) | Los Angeles, California |
| Publisher (of the digital version) | University of Southern California. Libraries |
| Provenance | Electronically uploaded by the author |
| Type | texts |
| Legacy record ID | usctheses-m3077 |
| Rights | Tiwana, Shahzad Majeed |
| Repository name | Libraries, University of Southern California |
| Repository address | Los Angeles, California |
| Repository email | http://www.usc.edu/isd/libraries/services/ask_a_librarian/email/ |
| Filename | etd-tiwana-3366 |
| Archival file | uscthesesreloadpub_Volume40/etd-tiwana-3366.pdf |
Description
| Title | Page 1 |
| Full text | METHODS FOR IMPROVING SEARCH ON U.S. PATENT DOCUMENTS by Shahzad Majeed Tiwana A Dissertation Presented to the FACULTY OF THE USC GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (COMPUTER SCIENCE) May 2010 Copyright 2010 Shahzad Majeed Tiwana |
Comments
Post a Comment for Page 1
