Patent Assignment Data Restructure

From edegan.com
Jump to navigation Jump to search


Project
Patent Assignment Data Restructure
Project logo 02.png
Project Information
Has title Patent Data Restructure
Has owner Marcela Interiano, Sonia Zhang
Has start date 201701
Has deadline date
Has keywords Patent, Data
Has project status Subsume
Does subsume Patent Data, Patent Data Cleanup - June 2016, Patent Data Extraction Scripts (Tool), USPTO Bulk Data Processing
Subsumed by: Reproducible Patent Data
Has sponsor McNair Center
Has project output Data
Copyright © 2019 edegan.com. All Rights Reserved.

In order to restructure the current patent dataset, the data requires rigorous cleaning. The primary areas for improvement are:

1. Clean ptoassignment table to unique keys.
2. Clean ptoproperties to remove nonutility patents. The patent numbers currently include:
  • 7 digit patent numbers
  • application numbers
  • unknown numbers that cannot be matched to patent numbers in the patent table
20090108066
20100007288
20090108066
20100110022
  • Design and Reissue patents ('%D%' or '%RE%')
  • alphanumeric character strings
3. Restructure address information in ptoassignee table to extract meaningful information
4. Verify that cleaned patent documentids correspond to patent numbers or application numbers in the patent table
5. Restructure address information in ptoassignment table
6. Transform structure of the dataset

Semester Plan

The final deliverable for the semester is a table with the following structure:

Reel No | Frame No | Invention Title | Filing Date | Patent No | Application No | Publication No | Match By

The schema of the table will be:

 Column          |          Type         | Modifiers
-----------------+-----------------------+-----------
Reel No          |        integer        |
Frame No         |        integer        |
Invention Title  | character varying(500)|
Filing Date      |          date         |
Patent No        |        integer        |
Application No   |        integer        |
Publication No   |        integer        |
Match By         |        integer        |

Currently the pto tables contain varying identifiers for one invention title that is involved in a reassignment. The purpose of the table is to have each identifier for an invention title listed in a single row, making it easy to track the invention through various transactions. The Match By column will inform the user which identifier should be used to match to other tables in the patent database.

Data Cleanup Progress

The tables that currently comprise the assignment data are as follows.

                Table "public.ptoassignment"
         Column          |          Type          | Modifiers 
-------------------------+------------------------+-----------
reelno                   | integer                | 
frameno                  | integer                | 
last_update_date         | date                   | 
purge_indicator          | character varying(2)   | 
recorded_date            | date                   | 
correspondent_name       | character varying(500) | 
correspondent_address_1  | character varying(500) | 
correspondent_address_2  | character varying(500) | 
correspondent_address_3  | character varying(500) | 
correspondent_address_4  | character varying(500) | 
conveyance_text          | character varying(500) |
Table "public.ptoassignee"
  Column   |          Type          | Modifiers 
-----------+------------------------+-----------
reelno     | integer                | 
frameno    | integer                | 
name       | character varying(500) | 
addrline1  | character varying(500) | 
addrline2  | character varying(500) | 
city       | character varying(500) | 
state      | character varying(500) | 
country    | character varying(500) | 
postcode   | character varying(80)  | 
Table "public.ptoassignor"
     Column     |          Type          | Modifiers 
----------------+------------------------+-----------
reel_no         | integer                | 
frame_no        | integer                | 
assignor_name   | character varying(500) | 
execution_date  | date                   | 
  Table "public.ptoproperty"
     Column      |          Type          | Modifiers 
-----------------+------------------------+-----------
reelno           | integer                | 
frameno          | integer                | 
documentid       | character varying(20)  | 
country          | character varying(500) | 
kind             | character varying(3)   | 
filingdate       | date                   | 
invention_title  | character varying(500) | 
Table "public.ptopatentfile"
         Column         |         Type          | Modifiers 
------------------------+-----------------------+-----------
reel_no                 | integer               | 
frame_no                | integer               | 
action_key_code         | character varying(10) | 
uspto_transaction_date  | date                  | 
uspto_date_produced     | date                  | 
version                 | numeric               |

This section explains the series of steps that were taken to clean and to take note of problems in the data. Additionally, this section includes the codes for new tables that combine patent properties from different tables in the original assignment data.

Table Cleanup

The five original assignment tables contain duplicates that were deleted. Five new tables were made and used as templates for building new tables.

Ptoassignmentnd Table:

SELECT COUNT(*) FROM ptoassignment;
--8676322
DROP ptoassignmentnd; 
CREATE TABLE ptoassignmentnd AS
SELECT reelno, frameno, max(last_update_date) as last_update_date, purge_indicator, recorded_date, correspondent_name, 
correspondent_address_1, correspondent_address_2, correspondent_address_3, correspondent_address_4, conveyance_text 
FROM ptoassignment GROUP BY reelno, frameno, purge_indicator, recorded_date, correspondent_name, correspondent_address_1, 
correspondent_address_2, correspondent_address_3, correspondent_address_4, conveyance_text;
--6988575

Ptoassigneend Table:

SELECT COUNT(*) FROM ptoassignee;
--8983280
DROP ptoassigneend; 
CREATE TABLE ptoassigneend AS
SELECT DISTINCT reelno, frameno, name, addrline1, addrline2, city, state, country, postcode FROM ptoassignee; 
--7234001

Ptoassignornd Table:

SELECT COUNT(*) FROM ptoassignor;
--20062463
DROP ptoassignornd; 
CREATE TABLE ptoassinornd AS 
SELECT DISTINCT reel_no, frame_no, assignor_name, execution_date FROM ptoassignor;
--16126903

Ptoproperty_cleaned Table:

SELECT COUNT(*) FROM ptoproperty;
--65214396
DROP TABLE ptoproperty_cleaned;
CREATE TABLE ptoproperty_cleaned AS
SELECT DISTINCT reelno, frameno, documentid, country, kind, filingdate, invention_title
FROM ptoproperty; 
--8696149

Ptopatentfilend Table:

SELECT COUNT(*) FROM ptopatentfile;
--8676317
DROP ptopatentfilend; 
CREATE TABLE ptopatentfilend AS 
SELECT DISTINCT reel_no, frame_no, action_key_code, uspto_transaction_date, uspto_date_produced, version FROM ptopatentfile; 
--7159725

Patent Number Cleanup

The goal is to only have assignment records on utility patents. The patents in ptoproperty include alphanumerics, which represent reissue and design patents as well as mistakes in the data input. Additionally, the documentids include application numbers or ids and publication numbers. The ptoproperty table stores the patent ids as character strings.

First the duplicates were dropped from the ptoproperty table creating ptoproperty_cleaned.

SELECT COUNT(*) FROM (SELECT DISTINCT * FROM ptoproperty) As T; 
--27266638
 
SELECT COUNT(*) FROM ptoproperty_cleaned;
--27266638

Next, the Reissue and Design patents were removed.

SELECT COUNT(*) FROM ptoproperty WHERE documentid LIKE 'RE%';
--38512
SELECT COUNT(*) FROM ptoproperty WHERE documentid LIKE 'D%';
--1128247

Finally, all letters were removed from the data, resulting in the final version of ptoproperty_cleaned.

The ptoproperty_cleaned table contains application numbers, publication numbers, and patent numbers. The patents may also have two distinct publication numbers based on the year in which it was published. Based on length of documentid, the three types of id numbers were separated into three separate tables.

DROP TABLE ptoproperty_patent; 
CREATE TABLE ptoproperty_patent AS 
SELECT * FROM ptoproperty_cleaned WHERE length(documentid) = 7; 
--8696149
ALTER TABLE ptoproperty_patent RENAME COLUMN documentid TO patentno;
DROP TABLE ptoproperty_app; 
CREATE TABLE ptoproperty_app AS
SELECT * FROM ptopropertynd WHERE length(documentid) = 8; 
--11577028
ALTER TABLE ptoproperty_app RENAME COLUMN documentid TO appno;
DROP TABLE ptoproperty_pub; 
CREATE TABLE ptoproperty_pub AS 
SELECT * FROM ptopropertynd WHERE length(documentid) = 11; 
--6217864
ALTER TABLE ptoproperty_pub RENAME COLUMN documentid TO pubno;

All three tables contain the following information and differ only on the type of patent id.

            Table "public.ptoproperty_patent"
     Column      |          Type          | Modifiers 
-----------------+------------------------+-----------
 reelno          | integer                | 
 frameno         | integer                | 
 documentid      | character varying(20)  | 
 country         | character varying(500) | 
 kind            | character varying(3)   | 
 filingdate      | date                   | 
 invention_title | character varying(500) | 

Ptotracking Tables

The purpose of the ptotracking tables is to track the ownership of patents based on the update date and filing dates of the assignment. These tables can be used to add missing information or track further refined tables such as ptoproperty_patent.

Ptotracking takes the reelno, frameno, and documentid key from the ptoproperty_cleaned table and joins the update dates and recorded dates corresponding to the transactions.

DROP TABLE ptotracking;
CREATE TABLE ptotracking AS 
SELECT M1.reelno, M1.frameno, M1.documentid, M2.last_update_date, M2.recorded_date
FROM ptoproperty_cleaned M1, ptoassignmentnd M2
WHERE (M1.reelno = M2.reelno) AND (M1.frameno = M2.frameno); 
--8699074

In making the ptotracking table, it is important to note that the classification of documentids as B1 and B2 causes duplicates in the entries. B1 and B2 classifications mean that the patent was granted with and without a published application.

Ptotracking2 adds the assignee to the transaction, allowing the user to track ownership of the entity and of the patent.

DROP TABLE ptotracking2;
CREATE TABLE ptotracking2 AS 
SELECT M1.reelno, M1.frameno, M1.documentid, M2.name, M1.last_update_date, M1.recorded_date
FROM ptotracking M1, ptoassigneend M2 
WHERE (M1.reelno = M2.reelno) AND (M1.frameno = M2.frameno);
--9613927

The document ids in the PTO assignment data had not yet been verified as matching to the main patent table in our database (psql patent). The document ids in the PTO assignment data are stored as character strings whereas the patents in the patent table are stored as integers. Unlike the ptoprpoperty_cleaned table, all patent numbers in the patent table are unique.

The following two tables were made in order to verify that the documentids in the ptoproperty_cleaned table match to the patent table.

DROP TABLE edcheck;  
CREATE TABLE edcheck AS 
SELECT CAST (documentid AS INT) FROM ptotracking2; 
SELECT COUNT(DISTINCT documentid) FROM edcheck; 
--2343765
DROP TABLE edcheck2; 
CREATE TABLE edcheck2 AS 
SELECT M1.documentid, M2.patent
FROM edcheck M1, patent M2 
WHERE (M1.documentid = M2.patent); 
--2238305
DROP TABLE edcheck; 
CREATE TABLE edcheck AS 
SELECT DISTINCT documentid FROM ptotracking2; 
--2343765
DROP TABLE edcheck2; 
CREATE TABLE edcheck2 AS 
SELECT CAST(documentid AS INT)FROM edcheck; 
--2343765
DROP TABLE edcheck3; 
CREATE TABLE edcheck3 AS 
SELECT M1.documentid, M2.patent FROM edcheck2 M1, patent M2 WHERE M1.documentid = M2.patent; 
--2238305

Based on the iterations of these tables, we could conclude that our original patent data forms the majority of the patents undergoing reassignments or transactions.

US ONLY Patent Assignee Table

Note: Table made for Julia by Marcela Interiano

Table made in the patent database using the USPTO assignment data.

The first step was to include the last_update_date in with the data from the ptoproperty table. The ptoproperty table contains only the filing date, which is not useful as we are looking for the current patent holders. The table ptoproperty_patent was used for the patent numbers as this table was cleaned to include only patent numbers, no application or publication numbers.

DROP TABLE ptoproperty_patent_update; 
CREATE TABLE ptoproperty_patent_update AS 
SELECT M1.reelno, M1.frameno, M1.patentno, M2.last_update_date 
FROM ptoproperty_patent M1, ptotracking2 M2 
WHERE (M1.reelno = M2.reelno) AND (M1.frameno = M2.frameno) AND (M1.patentno = M2.documentid); 

Next, the minimum update date was taken, dropping any repetitions or later dates for the same patent assignee.

DROP TABLE ptoproperty_patent_minupdate; 
CREATE TABLE ptoproperty_patent_minupdate AS 
SELECT reelno, frameno, patentno, min(last_update_date) FROM ptoproperty_patent_update GROUP BY reelno, frameno, patentno, last_update_date; 

The US only assignee table was used to construct the final table.

DROP TABLE ptoassignee_us_patent; 
CREATE TABLE ptoassignee_us_patent AS 
SELECT M1.reelno, M1.frameno, M2.name, M1.patentno, M1.last_update_date 
FROM ptoproperty_patent_minupdate M1, ptoassigneend_us_cleaned M2
WHERE (M1.reelno = M2.reelno) AND (M1.frameno = M2.frameno);

The table below was made to join through using Sonia's zip codes for the ptoassignee data to get patent numbers from reelno and frameno.

DROP TABLE ptoassignee_us_distinct; 
CREATE TABLE ptoassignee_us_distinct AS 
SELECT DISTINCT reelno, frameno, patentno 
FROM ptoassignee_us_patent 
GROUP  BY reelno, frameno, patentno; 
--5391413

The total number of distinct patent numbers in the ptoassignee data for only US assignees is 2345763.

SELECT COUNT(*) FROM (SELECT DISTINCT patentno FROM ptoassignee_us_patent) AS P;
--2345763

Current Assignee using Recorded Date

Each assignment has three dates: filingdate, recorded_date, last_update_date. The filingdate corresponds to the filing of the assignment with the USPTO. The recorded_date is the date the transaction was recorded. The last_update_date is the date the USPTO verifies that the assignment still holds. In the ptoassignee_us_patent table, the last_update_date is used to find the current assignee.

Prior to Sonia's work with the ptoassignee table address data, the table ptoassignee_current was made using the most recent recorded_date. This method though is flawed given that additional transactions could have current previously that are still in effect as patents can have multiple assignees. These codes can be used for constructing similar tables using the address data Sonia has cleaned in the following sections of this project.

To begin with, the ptoproperty_patent table was cleaned to drop all duplicates. Then the table was matched with the assignee table.

DROP TABLE ptoassigneev1; 
CREATE TABLE ptoassigneev1 AS 
SELECT M1.reelno, M1.frameno, M1.documentid, M1.country, M1.filingdate, M2.last_update_date, 
M2.recorded_date 
FROM ptoproperty_patent2 M1, ptoassignmentnd M2 
WHERE (M1.reelno = M2.reelno) AND (M1.frameno = M2.frameno); 
DROP TABLE ptoassigneev2;
CREATE TABLE ptoassigneev2 AS 
SELECT M1.reelno, M1.frameno, M1.documentid, M2.name, M1.country, M1.last_update_date, M1.recorded_date, 
M2.addrline1, M2.addrline2, M2.city, M2.state, M2.postcode
FROM ptoassigneev1 M1, ptoassigneend M2
WHERE (M1.reelno = M2.reelno) AND (M1.frameno = M2.frameno); 
--9634942

Once all the location and address fields from the ptoassignee table have been added to the ptoproperty_patent fields, the max recorded_date was identified from the ptoassignee_patent table and from ptoassigneev2 for comparison.

DROP TABLE datecheck;
CREATE TABLE datecheck AS 
SELECT documentid, max(recorded_date) as recorded_date FROM ptoassignee_patent GROUP BY documentid; 
--2343765
DROP TABLE datecheck;
CREATE TABLE datecheck AS 
SELECT documentid, max(recorded_date) as recorded_date FROM ptoassigneev2 GROUP BY documentid; 
--4374885
DROP TABLE ptoassignee_patent; 
CREATE TABLE ptoassignee_patent AS
SELECT M1.reelno, M1.frameno, M1.documentid, M1.name, M1.last_update_date, M1.recorded_date, M2.addrline1, 
M2.addrline2, M2.city, M2.state, M2.country, M2.postcode
FROM ptotracking2 M1, ptoassigneend M2 WHERE (M1.reelno = M2.reelno) AND (M1.frameno = M2.frameno);
--16581236
DROP TABLE ptoassignee_current; 
CREATE TABLE ptoassignee_current AS 
SELECT M1.reelno, M1.frameno, M2.documentid, M2.recorded_date FROM ptoassignee_patent M1, datecheck M2 
WHERE (M1.documentid = M2.documentid)
AND (M1.recorded_date = M2.recorded_date); 
--6729698

A final version of the ptoassignee_current table was made using ptoassigneev2 given the larger pool of documentids included in the table by matching using documentid and recorded dates from datecheck.

DROP TABLE ptoassignee_current; 
CREATE TABLE ptoassignee_current AS  
SELECT M1.reelno, M1.frameno, M2.documentid, M1.name, M1.last_update_date, M2.recorded_date, M1.addrline1, 
M1.addrline2, M1.country, M1.city, M1.state, M1.postcode
FROM ptoassigneev2 M1, datecheck M2 
WHERE (M1.documentid = M2.documentid) AND (M1.recorded_date = M2.recorded_date); 
--4994869

These codes should be used to recreate this table using Sonia's updated address information.

Matching Application and Publication Numbers

The ptoproperty_cleaned documentids to verify the kind of different patents as specified in the ptoproperty tables.

First the table ptopropertynd was made, including only the distinct documentids in ptoproperty_cleaned.

DROP ptopropertynd; 
CREATE TABLE ptopropertynd AS 
SELECT DISTINCT * FROM ptoproperty; 
--27266638

By creating this table, I also address the duplications caused by the kind XO.

Final Table (name TBD)

CREATE TABLE patpub AS 
(SELECT ptoproperty_patent.reelno, ptoproperty_patent.frameno, ptoproperty_patent.invention_title,
ptoproperty_patent.patentno, ptoproperty_pub.pubno FROM ptoproperty_patent LEFT JOIN ptoproperty_pub 
ON 
(ptoproperty_patent.reelno = ptoproperty_pub.reelno) AND (ptoproperty_patent.frameno = 
ptoproperty_pub.frameno)
AND (ptoproperty_patent.invention_title = ptoproperty_pub.invention_title)
UNION 
SELECT ptoproperty_pub.reelno, ptoproperty_pub.frameno, ptoproperty_pub.invention_title,
ptoproperty_patent.patentno, ptoproperty_pub.pubno FROM ptoproperty_patent RIGHT JOIN ptoproperty_pub 
ON 
(ptoproperty_patent.reelno = ptoproperty_pub.reelno) AND (ptoproperty_patent.frameno = 
ptoproperty_pub.frameno) 
AND (ptoproperty_patent.invention_title = ptoproperty_pub.invention_title));
CREATE TABLE patapp AS 
(SELECT ptoproperty_patent.reelno, ptoproperty_patent.frameno, ptoproperty_patent.invention_title,
ptoproperty_patent.patentno, ptoproperty_app.appno FROM ptoproperty_patent LEFT JOIN ptoproperty_app 
ON 
(ptoproperty_patent.reelno = ptoproperty_app.reelno) AND (ptoproperty_patent.frameno = 
ptoproperty_app.frameno)
AND (ptoproperty_patent.invention_title = ptoproperty_app.invention_title)
UNION 
SELECT ptoproperty_app.reelno, ptoproperty_app.frameno, ptoproperty_app.invention_title,
ptoproperty_patent.patentno, ptoproperty_app.appno FROM ptoproperty_patent RIGHT JOIN ptoproperty_app 
ON 
(ptoproperty_patent.reelno = ptoproperty_app.reelno) AND (ptoproperty_patent.frameno = 
ptoproperty_app.frameno) 
AND (ptoproperty_patent.invention_title = ptoproperty_app.invention_title));

Restructure Address Information (First Stage)

Note: This section was worked on by Sonia Zhang

The dbase is patent.

The table is ptoassigneend.

SQL code and other things are in:

E:/McNair/Projects/PatentAddress/RestructureAddressInfo(First Stage).sql

IO files are on the dbase server in:

Z:/PatentAddress

Introduction

Currently done:

  • Five features (addrline1, addrline2, city, state, postcode) in the table contain address information.
  • Features addrline1, addrline2 and city are not cleaned. They can have suite, street, city, state and postcode information, or any combination of these
  • The first object of this project is to extract postcode, state and city information from the three features above (Section 2.2.2).
  • Then, we summarize the postcode, state and city information in the original table and those extracted from addresses to generate only one postcode, state and city for each record (Section 2.2.3).
  • For now, we only focus on American patents.

Extract Address Information

So far these instructions apply to:

  • Records identified as US (by zip code or country feature) ~3.5m records
  • Those that well-formatted zip code (5-4) in addrline1, addrline2, and city (not postcode, etc) ~5,000 records

To get the record identified as US, we make the table:

  • ptoassigneend_allus

This table contains all the U.S. patents extracted from ptoassigneend table. The rule to generate this table is 'country = 'UNITED STATES' ' if any of the features contains U.S. postcode.

Noise exists because of postcode errors.

The ptoassigneend_allus table may miss some U.S. patents which lose postcode records.

SQL code:

CREATE TABLE ptoassigneend_allus AS
SELECT * FROM ptoassigneend
WHERE city ~* '.*\d{5}[-]\d{4}.*' OR addrline1 ~* '.*\d{5}[-]\d{4}.*' OR
addrline2 ~* '.*\d{5}[-]\d{4}.*' OR postcode ~* '.*\d{5}[-]\d{4}.*' OR
country ~* 'UNITED STATES';

SELECT 3572682
      Table "public.ptoassigneend_allus"
 Column   |          Type          | Modifiers
----------| -----------------------|-----------
reelno    | integer                |
frameno   | integer                |
name      | character varying(500) |
addrline1 | character varying(500) |
addrline2 | character varying(500) |
city      | character varying(500) |
state     | character varying(500) |
country   | character varying(500) |
postcode  | character varying(80)  |

Postcode

U.S. postcode should follow the pattern five digits - four digits. In this way, U.S. patents can be extracted by searching for postcode with regular expression.

Note: It is too risky to extract just 5 digit postcodes as these might street numbers, or other countries postcodes.

'(^|\s)\d{5}-\d{4}($|\s)'

For example,

                city                  | postcode_extracted
--------------------------------------|-------------------
NEW YORK, NY 10022-3201               | 10022-3201
BEAVERTON, OREGON 97005-6453          | 97005-6453
SANTA BARBARA, CA 93130-3003          | 93130-3003
NEW YORK NY 10022-3201                | 10022-3201
SUNNYVALE, CA 94088-3453              | 94088-3453
94088-3470                            | 94088-3470
CS 46510-35065 RENNES CEDEX           | 46510-3506
NEW YORK, NY 10013-2412               | 10013-2412
OALKLAND, CA 94612-3550               | 94612-3550
OXFORD CT 06483-1011                  | 06483-1011

The extracted postcode is in table ptoassigneend_us_extracted.

SQL code is in:

E:/McNair/Projects/PatentAddress/RestructureAddressInfo(First Stage).sql

State

The following patterns can be used to extract state information.

Note: a,b,c are to be run on city, addrline1, and addrline2. We use city as an example below.

a. ', State Postcode'

The state and postcode are always together, separated by a space. We can extract state information with regular expression

'([,]|[.])\s\w{2,}\s{0,}\w{0,}\s{1,}\d{5}[-]\d{4}' 

SQL Code:

WHEN city ~* '([,]|[.])\s\w{2,}\s{0,}\w{0,}\s{1,}\d{5}[-]\d{4}'
THEN LTRIM(RTRIM(replace(regexp_replace(SUBSTRING(city, '[,]\s\w{2,}\s{0,}\w{0,}\s{1,}\d{5}[-]\d{4}'), '\d{5}-\d{4}', ),',',)))

Examples:

                city                  |  state_city
--------------------------------------|-----------
NEW YORK, NY 10022-3201               | NY
BEAVERTON, OREGON 97005-6453          | OREGON
SANTA BARBARA, CA 93130-3003          | CA
SUNNYVALE, CA 94088-3453              | CA
NEW YORK, NY 10013-2412               | NY
OALKLAND, CA 94612-3550               | CA
SANTA CLARA, CA 95052-8090            | CA
PEORIA, IL  61629-6490                | IL
MIDVALE, UTAH 84047-1408              | UTAH
OAKLAND, CA 94612-3550                | CA
HARRISBURG, PA 17105-3608             | PA
ROCHESTER, NY  14650-2201             | NY
NEW YORK, NY  10013-2412              | NY
HOUSTON, TEXAS 77256-6571             | TEXAS
BROOKINGS, SOUTH DAKOTA 57006-0128    | SOUTH DAKOTA 

b. '\s State(abbreviation) Postcode'

'(^|\s)\w{2}\s{1}\d{5}[-]\d{4}'

SQL code:

WHEN city ~* '(^|\s)\w{2}\s{1}\d{5}[-]\d{4}' 
THEN regexp_replace(SUBSTRING(city, '\w{2}\s{1,}\d{5}[-]\d{4}'), '\d{5}-\d{4}', ) 

Examples:

NEW YORK NY 10022-3201    |NY
WAUKEGAN IL 60085-2195    |IL

c. 'D.C.'

'D[.]C[.]\s\d{5}-\d{4}'

SQL code:

WHEN city ~* 'D[.]C[.]\s\d{5}-\d{4}'
THEN 'D.C.'

d. 'A CORP.* OF State'

This pattern is not reliable. When 'addrline1' looks this way, 'addrline2' always provide more detailed address information than 'addrline1'. Besides, a great part of state info extracted from 'A CORP.* OF [State]' doesn't match the state extracted from detailed 'addrline2'. In this way, we discard this pattern.

Examples:

addrline1                                                                                | addrline2
-----------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------
A CORP. OF NEW YORK                                                                      | ONE LINCOLN FIRST SQUARE, ROCHESTER, NEW YORK, NY 14601-0054
A CORPORATION OF NY                                                                      | 550 MADISON AVENUE, NEW YORK, NY 10022-3201
A NON-PROFIT CORP. OF WASHINGTON                                                         | 4225 ROOSEVELT WAY NE, STE. 303 SEATTLE, WA 98105-6099
A CORP. OF MN                                                                            | 2611 NORTH SECOND ST.,                                      MINNEAPOLIS, MN 55411-1633
A CORPORATION OF DE                                                                      | 612 WHEELER'S FARM RD.                                      MILFORD, CT 06460-8719
A CORP. OF WA.                                                                           | 411 FIRST AVENUE SOUTH                                      SEATTLE, WASHINGTON 98104-2871
A CORPORATION OF OREGON                                                                  | ONE BOWERMAN DRIVE, BEAVERTON, OREGON 97005-6453
A CORPORATION OF                                                                         | ONE BOWERMAN DRIVE BEAVERTON, OR 97005-6453
A CORP. OF INDIANA                                                                       | 4810 TECUMSEH LANE, POST OFFICE BOX 15190                   EVANSVILLE, INDIANA 47716-0190
A CORPORATION OF PA                                                                      | 7200 SUTER ROAD, COOPERSBURG, PA 18036-1299
A CORP. OF DE                                                                            | SCARBORO & BEAR CREEK ROAD, P.O. BOX 2009                   OAK RIDGE, TENNESSEE 37831-8014

If you're interested, the SQL code is

WHEN addrline1 ~* 'CORP.*OF(\s|$)'
THEN LTRIM(RTRIM(SUBSTRING(regexp_replace(SUBSTRING(addrline1, 'CORP.*OF.*'),'CORP.*OF', ), 1, 18)))
  • Summary

The extracted state records are stored in the table ptoassigneend_us_extracted.

SQL code is in:

E:/McNair/Projects/PatentAddress/RestructureAddressInfo(First Stage).sql

City

The following patterns can be used to extract city information.

a. '\s{2,} CityName [,] State Postcode'

SQL code:

CASE WHEN addrline1 ~* '\s{2,}\w{1,}\s{0,}\w{0,}\s{0,}[,]\s{0,}\w{1,}\s{0,}\w{0,}\s{0,}\d{5}[-]\d{4}' 
THEN LTRIM(RTRIM(REPLACE(SUBSTRING(SUBSTRING(addrline1, '\s{2,}\w{1,}\s{0,}\w{0,}\s{0,}[,]\s{0,}\w{1,}\s{0,}\w{0,}\s{0,}\d{5} [-]\d{4}'),'.*[,]'),',',)))

Examples (in any of addrline1, addrline2, and city):

source                                                                                          |extracted
------------------------------------------------------------------------------------------------|---------------------
800 CHRYSLER DR. EAST                                       AUBURN HILLS, MICHIGAN  48326-2757  |AUBURN HILLS
550 MADISON AVENUE                                          NEW YORK, NEW YORK 10022-3201       |NEW YORK
P.O. BOX 15439                                              WILMINGTON, DE  19850-5439          |WILMINGTON

Some noise exists (just a little).

1313 N. MARKET STREET                                       HERCULES PLAZAWILMINGTON, DE  19894-0001

b. '[,]\s{1,} CityName [,] State Postcode'

SQL code:

CASE WHEN addrline1 ~* '\s{2,}\w{1,}\s{0,}\w{0,}\s{0,}[,]\s{0,}\w{1,}\s{0,}\w{0,}\s{0,}\d{5}[-]\d{4}' 
THEN LTRIM(RTRIM(REPLACE(SUBSTRING(SUBSTRING(addrline1, '\s{2,}\w{1,}\s{0,}\w{0,}\s{0,}[,]\s{0,}\w{1,}\s{0,}\w{0,}\s{0,}\d{5}[-]\d{4}'),'.*[,]'),',',)))

Example:

source                                                                     |extracted
---------------------------------------------------------------------------|---------------------
920 DISC DRIVE, SCOTTS VALLEY, CA 95067-0360                               |SCOTTS VALLEY
550 MADISON AVENUE, NEW YORK, NY 10022-3201                                |NEW YORK
BALLSTON TOWER ONE 800 NORTH QUINCY STREET, ARLINGTON, VA   22217-5660     |ARLINGTON

c. 'CityName [,] State Postcode' (no leading spaces)

SQL code:

WHEN addrline1 ~* '^\w{1,}\s{0,}\w{0,}\s{0,}[,]\s{0,}\w{1,}\s{0,}\w{0,}\s{0,}\d{5}[-]\d{4}'
THEN REPLACE(SUBSTRING(SUBSTRING(addrline1, '^\w{1,}\s{0,}\w{0,}\s{0,}[,]\s{0,}\w{1,}\s{0,}\w{0,}\s{0,}\d{5}[-]\d{4}'), '.*[,]'),',', )

Examples:

PHILADELPHIA, PA 19104-3147   |PHILADELPHIA
ROCHESTER, NY 14650-2201      |ROCHESTER

d. 'CityName State(abbreviation) Postcode' (no leading spaces)

SQL code:

WHEN addrline1 ~* '^\w{1,}\s{0,}\w{0,}\s{0,}\w{2}\s{0,}\d{5}[-]\d{4}'
THEN regexp_replace(SUBSTRING(addrline1, '^\w{1,}\s{0,1}\w{0,}\s{1,}\w{2}\s{0,}\d{5}[-]\d{4}'), '\w{2}\s\d{5}[-]\d{4}', )

Examples:

TARRYTOWN NY 10591-6706       |TARRYTOWN

e. Special cases

Examples:

 BATON, ROUGE, LA 70809-4562 ('BATON ROUGH' is separated by a comma)          
 ST. PAUL, MIN 55133-3427 ('ST. PAUL' contains a dot)
 QUINCY STREETARLINGTON, VA 22217-5660 (no space between street and city name :(

SQL code:

CASE WHEN addrline1 ~* 'BATON[,] ROUGE[,]\s{1}LA'
THEN 'BATON ROUGE'
WHEN addrline1 ~* 'ST[.]\s{1}PAUL'
THEN 'ST. PAUL'
WHEN addrline1 ~* 'ARLINGTON[,]\s{1}VA'
THEN 'ARLINGTON'

Noise:

  • CityName State (full name) Postcode' (no leading spaces)
NEW YORK NEW YORK 10022-3201  

This pattern can't be identified because of much noise.

  • 'CityName Postcode' (no leading spaces)
LITTLE ELM  75068-3787    
OAK RIDGE  37831-6498                      

This pattern can't be identified because of the noise:

MASSACHUSETTS 02780-7319   ('State'+'Postcode')
  • no space between street and city name :(
BOX 87703CHICAGO, IL  60680-0703 

SQL code is in:

E:/McNair/Projects/PatentAddress/RestructureAddressInfo(First Stage).sql

Output (Tables)

  • ptoassigneend_us_extracted

Contain all the original features as well as city, state and postcode info extracted from features addrline1, addrline2 and city. See Section 2.2.1 for details.

     Table "public.ptoassigneend_us_extracted"
    Column     |          Type          | Modifiers
---------------|------------------------|-----------
reelno         | integer                |
frameno        | integer                |
name           | character varying(500) |
addrline1      | character varying(500) |
addrline2      | character varying(500) |
city           | character varying(500) |
state          | character varying(500) |
country        | character varying(500) |
postcode       | character varying(80)  |
postqcode_city | text                   |
postcode_addr1 | text                   |
postcode_addr2 | text                   |
state_city     | text                   |
state_addr1    | text                   |
state_addr2    | text                   |
city_addr1     | text                   |
city_addr2     | text                   |
city_city      | text                   |
  • postcode_city is the postcode extracted from 'city'
  • postcode_addr1 is the postcode extracted from 'addrline1'
  • postcode_addr2 is the postcode extracted from 'addrline2'
  • state_city is the state name extracted from 'city'
  • state_addr1 is the state name extracted from 'addrline1'
  • state_addr2 is the state name extracted from 'addrline2'
  • city_city is the city name extracted from 'city'
  • city_addr1 is the city name extracted from 'addrline1'
  • city_addr2 is the city name extracted from 'addrline2'

SQL code is in:

E:/McNair/Projects/PatentAddress/RestructureAddressInfo(First Stage).sql

Clean Address Info (Master Table)

Introduction

As mentioned in Section 2.2.2, city, state and postcode info are extracted from 'addrline1', 'addrline2' and 'city'. Original table also contains 'postcode', 'city' and 'state'. In this way, we have four candidates for city, state and postcode.

Info extracted from addresses and that in the original table are not necessarily consistent.

The object of this section is to pick out the best postcode, city and state for each record, and create a Master Table with original features and cleaned postcode, city, and state.

Postcode

Reminder: 'postcode_city' is the postcodes extracted from 'city'; 'postcode_addr1' is the postcodes extracted from 'addrline1'; 'postcode_addr2' is the postcodes extracted from 'addrline2'.

The 'postcode_city', 'postcode_addr1' and 'postcode_addr2' are all consistent!

Examples:

postcode_addr1 | postcode_addr2
---------------|---------------
90401-1708     | 90401-1708
37831-8243     | 37831-8243

The issue is the inconsistency between 'postcode' and ('postcode_city', 'postcode_addr1' and 'postcode_addr2').

  • Inconsistency between 'postcode' and 'postcode_addr1'

Case 1: 'postcode_addr1' beats 'postcode' because 'addrline1' is detailed. For example:

                                                                        addrline1                                                                          | postcode_addr1 | postcode
-----------------------------------------------------------------------------------------------------------------------------------------------------------|----------------|-----------
P.O. BOX 6 / 83707-0006                                                                                                                                    | 83707-0006     | 83716
BLDG. C01, M.S. A126 P.O. BOX 80028 LOS  ANGELES, CA 90080-0028                                                                                            | 90080-0028     | 90045
P. O. BOX 1407, HOUSTON, TEXAS 77251-1407                                                                                                                  | 77251-1407     | 77042
841 3RD, LONGVIEW, WASHINGTON 98632-0189                                                                                                                   | 98632-0189     | 95404
2210 W. OAKLAWN, 72762-6999                                                                                                                                | 72762-6999     | 72765-2020
BLDG. CO1, M.S. A126, P.O. BOX 80028 LOS ANGELES, CA  90080-0028                                                                                           | 90080-0028     | 90045
P.O. BOX 6 / 83707-0006                                                                                                                                    | 83707-0006     | 83716
  • Inconsistency between 'postcode' and 'postcode_addr2'

Case 2: 'postcode_addr2' beats 'postcode' because 'addrline2' is detailed.

Example:

                                              addrline2                                               | postcode_addr2 | postcode
------------------------------------------------------------------------------------------------------|----------------|----------
P.O. BOX 6 / 83707-0006                                                                               | 83707-0006     | 83716-9632
P.O. BOX 5800 - MS0161, ALBUQUERQUE, NEW MEXICO 87185-0161                                            | 87185-0161     | 87123-0161
P.O. BOX 674412, HOUSTON, TEXAS 77267-4412                                                            | 77267-4412     | 77002
WASHINGTON, DC  20250-1400                                                                            | 20250-1400     | 61604
BOX 10076, HATTIESBURGH, MISSISSIPPI  39406-0076                                                      | 39406-0076     | 39402
P.O. BOX 674412, HOUSTON, TEXAS 77267-4412                                                            | 77267-4412     | 77032
P.O. BOX 6 / 83707-0006                                                                               | 83707-0006     | 83716-9632
P.O. BOX 6 / 83707-0006                                                                               | 83707-0006     | 83716-9632
P.O. BOX 27115-4191                                                                                   | 27115-4191     | 27105-4191
P.O. BOX 674412, HOUSTON, TX 77267-4412                                                               | 77267-4412     | 77002

Besides, I randomly picked some records and googled address and postcode. The results support 'postcode_addr2'.

Examples:

Micron Technology, Inc.
8000 South Federal Way
Post Office Box 6
Boise 83707-0006
USA
Tel: Ph: 208-368-4000
WebsterBank, NA
50 Kennedy Plaza
11th Floor Suite 110 (Mail Stop: PR-105)
Providence, RI 02903
  • Inconsistency between 'postcode' and 'postcode_city'

Case 3: 'postcode_city' beats 'postcode'.

        city          |  state   | postcode_city | postcode
----------------------|----------|---------------|----------
HAUPPAUGE, 11788-8847 | NEW YORK | 11788-8847    | 78759
  • Output

SQL code is in:

E:/McNair/Projects/PatentAddress/RestructureAddressInfo(First Stage).sql

For records of which 'addrline1', 'addrline2' and 'city' don't contain postcode info, just clean the 'postcode' as the 'postcode_cleaned'

All the cleaned postcodes for U.S. patents are stored in ptoassigneend_us_cleaned (see feature postcode_cleaned).

State

Reminder: 'state_city' is the states extracted from 'city'; 'state_addr1' is the states extracted from 'addrline1'; 'state_addr2' is the states extracted from 'addrline2'.

The 'state_city', 'state_addr1' and 'state_addr2' are consistent.

Examples:

state_addr2 | state_city
------------|-----------
ILLINOIS    | ILLINOIS
IL          | IL
IL          | IL
CT          | CT
MN          | MN
CA          | CA

One reason for the inconsistency between 'state' and ('state_city', 'state_addr1' and 'state_addr2') is the co-existence of abbreviation and full form. So they are the same actually. Otherwise, 'state_city', 'state_addr1' and 'state_addr2' beat 'state' because they are detailed.

For records of which 'addrline1', 'addrline2' and 'city' don't contain state info, just take 'state' as the 'state_cleaned'.

  • Output

SQL code is in:

E:/McNair/Projects/PatentAddress/RestructureAddressInfo(First Stage).sql

All the cleaned states for U.S. patents are stored in ptoassigneend_us_cleaned (see feature state_cleaned).

Note: We might want to convert state names to standard codes.

City

Reminder: 'city_city' is the city info extracted from 'city'; 'city_addr1' is the city info extracted from 'addrline1'; 'city_addr2' is the city info extracted from 'addrline2'.

The 'city_city', 'city_addr1' and 'city_addr2' are consistent.

Examples:

  city_addr2 |  city_city
-------------|--------------
BOISE        | BOISE
THOMASVILLE  | THOMASVILLE
CARROLLTON   | CARROLLTON
CINCINNATI   | CINCINNATI
CINCINNATI   | CINCINNATI
PEORIA       | PEORIA
OAK RIDGE    | OAK RIDGE
CARROLLTON   | CARROLLTON

'city_city', 'city_addr1' and 'city_addr2' beat 'city' because they are detailed.

For records of which 'addrline1', 'addrline2' and 'city' don't contain city info, just keep 'city' as the 'city_cleaned'.

  • Output

SQL code is in:

E:/McNair/Projects/PatentAddress/RestructureAddressInfo(First Stage).sql

All the cleaned cities for U.S. patents are stored in ptoassigneend_us_cleaned. (see feature city_cleaned)

Output (Table)

  • ptoassigneend_us_cleaned
        Table "public.ptoassigneend_us_cleaned"
      Column        |          Type          | Modifiers
--------------------|------------------------|-----------
reelno              | integer                |
frameno             | integer                |
name                | character varying(500) |
addrline1           | character varying(500) |
addrline2           | character varying(500) |
city                | character varying(500) |
state               | character varying(500) |
country             | character varying(500) |
postcode            | character varying(80)  |
postcode_addr1      | text                   |
postcode_addr2      | text                   |
postcode_city       | text                   |
state_addr1         | text                   |
state_addr2         | text                   |
state_city          | text                   |
city_addr1          | text                   |
city_addr2          | text                   |
city_city           | text                   |
postcode_cleaned    | text                   |
postcode_f5_cleaned | text                   |
state_cleaned       | text                   |
city_cleaned        | text                   |

Feature postcode_cleaned, postcode_f5_cleaned (first five digits), state_cleaned and city_cleaned are cleaned postcode, state and city info.

Functions to Simplify SQL Code

Extraction

  • Postcode

SQL function:

CREATE OR REPLACE FUNCTION ExtractPostcode(adr text) RETURNS text AS $$ 
	BEGIN
		RETURN SUBSTRING(adr, '\d{5}[-]\d{4}');
	END;
$$ LANGUAGE plpgsql;

Explanation:

adr is the feature from which you want to extract postcode.
  • State

SQL function:

CREATE OR REPLACE FUNCTION ExtractState(adr text) RETURNS text AS $$
    SELECT 
CASE WHEN adr ~* '([,]|[.])\s\w{2,}\s{0,}\w{0,}\s{1,}\d{5}[-]\d{4}'
THEN LTRIM(RTRIM(replace(regexp_replace(SUBSTRING(adr, '[,]\s\w{2,}\s{0,}\w{0,}\s{1,}\d{5}[-]\d{4}'), '\d{5}-\d{4}', ),',',)))
WHEN adr ~* '(^|\s)\w{2}\s{1}\d{5}[-]\d{4}' 
THEN regexp_replace(SUBSTRING(adr, '\w{2}\s{1,}\d{5}[-]\d{4}'), '\d{5}-\d{4}', ) 
WHEN adr ~* 'D[.]C[.]\s\d{5}-\d{4}'
THEN 'D.C.'
ELSE NULL END AS result;
$$ LANGUAGE SQL;

Explanation:

adr is the feature from which you want to extract state.
  • City

SQL function:

CREATE OR REPLACE FUNCTION ExtractCity(adr text) RETURNS text AS $$
   SELECT 
CASE WHEN adr ~* '\s{2,}\w{1,}\s{0,}\w{0,}\s{0,}[,]\s{0,}\w{1,}\s{0,}\w{0,}\s{0,}\d{5}[-]\d{4}' 
THEN LTRIM(RTRIM(REPLACE(SUBSTRING(SUBSTRING(adr, '\s{2,}\w{1,}\s{0,}\w{0,}\s{0,}[,]\s{0,}\w{1,}\s{0,}\w{0,}\s{0,}\d{5}[-]\d{4}'),'.*[,]'),',',)))
WHEN adr ~* '[,]\s{0,}\w{1,}\s{0,}\w{0,}\s{0,}[,]\s{0,}\w{1,}\s{0,}\w{0,}\s{0,}\d{5}[-]\d{4}'
THEN LTRIM(RTRIM(REPLACE(SUBSTRING(SUBSTRING(adr, '[,]\s{0,}\w{1,}\s{0,}\w{0,}\s{0,}[,]\s{0,}\w{1,}\s{0,}\w{0,}\s{0,}\d{5}[-]\d{4}'),'[,].*[,]'),',',)))
WHEN adr ~* '^\w{1,}\s{0,}\w{0,}\s{0,}[,]\s{0,}\w{1,}\s{0,}\w{0,}\s{0,}\d{5}[-]\d{4}'
THEN REPLACE(SUBSTRING(SUBSTRING(adr, '^\w{1,}\s{0,}\w{0,}\s{0,}[,]\s{0,}\w{1,}\s{0,}\w{0,}\s{0,}\d{5}[-]\d{4}'), '.*[,]'),',', )
WHEN adr ~* '^\w{1,}\s{0,}\w{0,}\s{0,}\w{2}\s{0,}\d{5}[-]\d{4}'
THEN regexp_replace(SUBSTRING(adr, '^\w{1,}\s{0,1}\w{0,}\s{1,}\w{2}\s{0,}\d{5}[-]\d{4}'), '\w{2}\s\d{5}[-]\d{4}', )
WHEN adr ~* 'BATON[,] ROUGE[,]\s{1}LA'
THEN 'BATON ROUGE'
WHEN adr ~* 'ST[.]\s{1}PAUL'
THEN 'ST. PAUL'
WHEN adr ~* 'ARLINGTON[,]\s{1}VA'
THEN 'ARLINGTON'
ELSE NULL END AS result;
$$ LANGUAGE SQL;

Explanation:

adr is the feature from which you want to extract city.

With these functions, we can generate ptoassigneend_us_extracted, which is the final output in Section 2.2.1, with the following code:

CREATE TABLE ptoassigneend_us_extracted AS
SELECT *, ExtractPostcode(city) postcode_city,
ExtractPostcode(addrline1) postcode_addr1,
ExtractPostcode(addrline2) postcode_addr2,
ExtractState(city) state_city,
ExtractState(addrline1) state_addr1,
ExtractState(addrline2) state_addr2,
ExtractCity(city) city_city,
ExtractCity(addrline1) city_addr1,
ExtractCity(addrline2) city_addr2
FROM ptoassigneend_missus;

All the SQL is in:

E:/McNair/Projects/PatentAddress/RestructureAddressInfo(First Stage).sql

Cleaning

  • Clean Postcode

SQL function:

CREATE OR REPLACE FUNCTION PostcodeClean (text,text,text,text) RETURNS text AS $$
    $pri1=$_[0];
    $pri2=$_[1];
    $pri3=$_[2];
    $postcode=$_[3];
    if ($pri1) {return $pri1;}
    if ($pri2) {return $pri2;}
    if ($pri3) {return $pri3;}
    if ($postcode) {
    	if ($postcode = ~ '\d{5}-\d{4}') {return $postcode}
    	if ($postcode = ~ '\d{5}') {return $postcode}
    }
    return undef;
$$ LANGUAGE plperl;

Explanation:

$pri1 is the feature with the highest priority, which in this case (see Section 2.2.2.1 is postcode_addr1)
$pri2 is the feature with the second highest priority, which in this case (see Section 2.2.2.1 is postcode_addr2)
$pri3 is the feature with the third highest priority, which in this case (see Section 2.2.2.1 is postcode_city)
$postcode is the feature postcode.
  • Clean State

SQL function:

CREATE OR REPLACE FUNCTION StateClean (text,text,text,text) RETURNS text AS $$
    $pri1=$_[0];
    $pri2=$_[1];
    $pri3=$_[2];
    $state=$_[3];
    if ($pri1) {return $pri1;}
    if ($pri2) {return $pri2;}
    if ($pri3) {return $pri3;}
    if ($state) {return $state;}
    return undef;
$$ LANGUAGE plperl;

Explanation:

$pri1 is the feature with the highest priority, which in this case (see Section 2.2.2.2 is state_addr1)
$pri2 is the feature with the second highest priority, which in this case (see Section 2.2.2.2 is state_addr2)
$pri3 is the feature with the third highest priority, which in this case (see Section 2.2.2.2 is state_city)
$state is the feature state.
  • Clean City

SQL function:

CREATE OR REPLACE FUNCTION CityClean (text,text,text,text) RETURNS text AS $$
    $pri1=$_[0];
    $pri2=$_[1];
    $pri3=$_[2];
    $city=$_[3];
    if ($pri1) {return $pri1;}
    if ($pri2) {return $pri2;}
    if ($pri3) {return $pri3;}
    if ($city) {return $city;}
    return undef;
$$ LANGUAGE plperl;

Explanation:

$pri1 is the feature with the highest priority, which in this case (see Section 2.2.2.3 is city_addr1)
$pri2 is the feature with the second highest priority, which in this case (see Section 2.2.2.3 is city_addr2)
$pri3 is the feature with the third highest priority, which in this case (see Section 2.2.2.3 is city_city)
$city is the feature city.

Issues

  • Inconsistency between 'addrline' and 'country'

Example:

                                          addrline2                                           |            city             | country
2882 SAND HILL ROAD                                         MENLO PARK, CALIFORNIA 94025-7022 | TOKYO                       | JAPAN
2801 CENTERVILLE ROAD, P.O. BOX 15439                       WILMINGTON, DE 19850-5439         | TOKYO                       | JAPAN
1-6, UCHISAIWAI-CHO 1-CHOME                                                                   | CHIYODA-KU, TOKYO           | JAPAN
MENLO PARK, CA 94025-7022                                                                     | TOKYO                       | JAPAN
                                                                                              | MINATO-KU, TOKYO 10585-8518 | JAPAN
1225 NORTH HIGHWAY 169,                                     MINNEAPOLIS, MINNESOTA 55441-5058 | TOKYO                       | JAPAN
2882 SAND HILL ROAD                                         MENLO PARK, CA 94025-7022         | TOKYO                       | JAPAN
3001 ORCHARD PARKWAY                                        SAN JOSE, CALIFORNIA 95134-2088   | TOKYO 107                   | JAPAN
3001 ORCHARD PARKWAY                                        SAN JOSE, CA  95134-2088          | MINATO-KU, TOKYO 107        | JAPAN
3001 ORCHARD PARKWAY                                        SAN JOSE, CALIFORNIA 95134-2088   | TOKYO 107                   | JAPAN
3001 ORCHARD PARKWAY                                        SAN JOSE, CA 95134-2088           | MINATO-KU, TOKYO 107        | JAPAN


  • The post code and zip regex for other countries besides U.S. can be found here:
http://stackoverflow.com/questions/578406/what-is-the-ultimate-postal-code-and-zip-regex

Example:

"US", "\d{5}([ \-]\d{4})?"
"CA", "[ABCEGHJKLMNPRSTVXY]\d[ABCEGHJ-NPRSTV-Z][ ]?\d[ABCEGHJ-NPRSTV-Z]\d"
"DE", "\d{5}"
"JP", "\d{3}-\d{4}"
"FR", "\d{2}[ ]?\d{3}"
  • The city feature needs to be standardized. For example, 'GRAND CAYMAN, CAYMAN ISLAND' and 'GRAND CAYMAN' indicate the same city.
  • Both state name and its abbreviation exist.

Restructure Address Information (Second Stage)

Note: This section was worked on by Sonia Zhang

The dbase is patent.

The table is ptoassigneend.

SQL code and other things are in:

E:/McNair/Projects/PatentAddress/RestructureAddressInfo(Second Stage).sql

IO files are on the dbase server in:

Z:/PatentAddress

To do

In no particular order:

  • Remove city, state, zip, country from addrline1 & addrline2 to get clean addrlines.
  • Maybe concatenate addrline1 and addrline to make addrline
  • Identify clean data (e.g. City that is a city, zip that is a zip, state that is a state)
    • By pattern, length, match to list
  • Try some more patterns, perhaps with a slightly higher false positive rate, on the remaining uncleaned data
    • Iterate!

Identify Clean Data

Output: ptoassigneend_us_identify0

As mentioned in Section 3, the ptoassigneend_us_extracted is clean. Copy all the records in ptoassigneend_us_extracted to ptoassigneend_us_identify0.

Store remaining records in ptoassigneend_us_temp.

Output: ptoassigneend_us_identify1

The following section works on ptoassigneend_us_temp.

Filter out records with city that is a city, zip that is a zip, state that is a state.

Note: The consistency between city and state or city and postcode is not checked in this section.

  • ptoassigneend_us_citylist

Copy clean city records in ptoassigneend_us_extracted to ptoassigneend_us_citylist (775).

Since the city list is not long, I briefly cleaned the list by hand, and updated the ptoassigneend_us_citylist (730).

  • zip that is a zip

Match the pattern 5d-4d or 5 digits.

  • state that is a state

Select distinct state records with

SELECT DISTINCT state FROM ptoassigneend_us_temp;

The output shows that all the records not null or not spaces are valid state names.

  • city that is a city

One method to identify clean city is to find city records that match ptoassigneend_us_citylist.

SQL Code:

SELECT city
FROM ptoassigneend_us_temp
WHERE city IN (
    SELECT citylist
    FROM ptoassigneend_us_citylist);
-- SELECT 2603422
  • The table ptoassigneend_us_identify1 stores records that meet all the requirements above: zip with 5-4 or 5 digits, state not null and not spaces, and city in ptoasigneend_us_citylist.

SQL Code:

CREATE TABLE ptoassigneend_us_identify1 AS
SELECT *
FROM ptoassigneend_us_temp
WHERE city IN (
    SELECT citylist
    FROM ptoassigneend_us_citylist) AND
    state IS NOT NULL AND state !=  AND
    postcode ~* '\d{5}';
SELECT 2511356

Store remaining records in ptoassigneend_us_temp2.

Output: ptoassigneend_us_identify2

Part of 'city' contains comma at the end. Remove comma, and then match 'city' with ptoassigneend_us_citylist.

SELECT *, replace(city, ',', ) clean_city
FROM ptoassigneend_us_temp2;
# SELECT 1055874

The output is ptoassigneend_us_identify2.

SQL code:

CREATE TABLE ptoassigneend_us_identify2 AS
SELECT *
FROM ptoassigneend_us_temp3
WHERE clean_city IN (
    SELECT citylist
    FROM ptoassigneend_us_citylist) AND
    state IS NOT NULL AND state !=  AND
    postcode ~* '\d{5}';    
# SELECT 14508

Store remaining data (excluding data in ptoassigneend_us_identify0, ptoassigneend_us_identify1 & ptoassigneend_us_identify2) in ptoassigneend_us_temp3.

Clean Address: more patterns

Table: ptoassigneend_us_temp3

SQL code is in E:\McNair\Projects\PatentAddress\RestructureAddressInfo(Second Stage).sql

Clean Postcode

Identifying five-digit postcode is risky because of the existence of P.O. BOX #, SUITE #, etc.

One option is to identify state and postcode together with the following SQL function:

CREATE OR REPLACE FUNCTION ExtractPostcode2(adr text) RETURNS text AS $$
    SELECT CASE WHEN (adr ~* '([,]|[.])\s\w{2,}\s{0,}\w{0,}\s{1,}\d{5}' OR 
adr ~* '(^|\s)\w{2}\s{1}\d{5}') AND 
NOT (adr ~* 'BO' OR adr ~* 'P[.]O') AND
NOT (adr ~* 'SUITE\s\d{5}')
THEN SUBSTRING(adr, '\d{5}')
ELSE NULL END AS result;
$$ LANGUAGE SQL;

Examples:

                                                                                                    addrline1                                                                                                     | substring
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+-----------
1650 WEST BIG BEAVER ROAD                                   TROY, MI 48084                                                                                                                                        | 48084
1114 AVE NY NY 10036                                                                                                                                                                                              | 10036
GLENDALE, CA  91204                                                                                                                                                                                               | 91204
14714 F. PERTHSHIRE 77079                                                                                                                                                                                         | 14714
314 N. JACKSON STREET, JACKSON 49201                                                                                                                                                                              | 49201
LAGUNA HILLS, CA 92653                                                                                                                                                                                            | 92653
1 ARAB, ALABAMA 35016                                                                                                                                                                                             | 35016
1205 SIXTH ST. SOUTHEAST 33907                                                                                                                                                                                    | 33907
767 FIFTH AVE.,                                             NEW YORK, NY 10153                                                                                                                                    | 10153

Noise still exists.

After extracting postcode, the following function is used to get clean postcode.

The priority is 'postcode' if it is '\d{5}', postcode_addr1, postcode_addr2, postcode_city

CREATE OR REPLACE FUNCTION PostcodeClean2 (text,text,text,text) RETURNS text AS $$
   $pri1=$_[0];
   $pri2=$_[1];
   $pri3=$_[2];
   $postcode=$_[3];
   if ($postcode) {return $postcode;}
   if ($pri1) {return $pri1;}
   if ($pri2) {return $pri2;}
   if ($pri3) {return $pri3;}
   return undef;
$$ LANGUAGE plperl;

The details and SQL function are in E:\McNair\Projects\PatentAddress\RestructureAddressInfo(Second Stage).sql

The output is table ptoassigneend_us_postex which include a new feature 'postcode_extracted'.

Clean 'city'

'city' is cleaned using following patterns.

  • Pattern 1: 'city' ~ 'city name, state ID'
  • Pattern 2: 'city' ~ 'city name, state postcode (5 digits)'
  • Pattern 3: 'city' ~ 'city name,'

The SQL function is:

CREATE OR REPLACE FUNCTION ExtractCity2(adr text) RETURNS text AS $$
    SELECT 
CASE WHEN adr ~* '[,]\s{1,}\w{2}(\s{1,}|$|[.]|[,])' OR
adr ~* '[,]\s{1}\w{1}[.]\w{1}[.]'
THEN REPLACE(SUBSTRING(adr, '.*[,]'),',',)
WHEN adr ~* '[,].*\d{5}$'
THEN REPLACE(SUBSTRING(adr, '.*[,]'),',',)
WHEN adr ~* '.*[,]$'
THEN REPLACE(adr, ',', )
ELSE adr END AS result;
$$ LANGUAGE SQL;

The details and SQL function are in E:\McNair\Projects\PatentAddress\RestructureAddressInfo(Second Stage).sql

The output is table ptoassigneend_us_postex2 which include a new feature 'city_extracted' and 'postcode_extracted'.

CREATE TABLE ptoassigneend_us_postex2 AS
SELECT *, ExtractCity2(city) city_extracted
FROM ptoassigneend_us_postex;

Identify Clean Data (Round Two)

A new list of clean city is extracted in Section 4.3.2. This list, combined with 'ptoassigneend_us_citylist', creates a new city list 'ptoassigneend_us_citylist2' which can be used to identify clean data.

Since the city list is not long, I briefly cleaned it by hand, and stored it in ptoassigneend_us_citylist2.

Output: ptoassigneend_us_identify3

Similar to Section 4.2, identify clean data that meets all the requirements: postcode_extracted with 5-4 or 5 digits, state not null and not spaces, and city_extracted in ptoasigneend_us_citylist2.

SQL Code:

CREATE TABLE ptoassigneend_us_identify3 AS
SELECT *
FROM ptoassigneend_us_postex2 
WHERE city_extracted IN (
   SELECT citylist
   FROM ptoassigneend_us_citylist2) AND
   state IS NOT NULL AND state !=  AND
   postcode_extracted ~* '\d{5}';
SELECT 664524
  • Output table: ptoassigneend_us_identify3
       Table "public.ptoassigneend_us_identify3"
      Column       |          Type          | Modifiers
-------------------|------------------------|-----------
reelno             | integer                |
frameno            | integer                |
name               | character varying(500) |
addrline1          | character varying(500) |
addrline2          | character varying(500) |
city               | character varying(500) |
state              | character varying(500) |
country            | character varying(500) |
postcode           | character varying(80)  |
postcode_extracted | text                   |
city_extracted     | text                   |

Remaining records are in table ptoassigneend_us_temp4.

Output: ptoassigneend_us_identify4

Some of the city records contain dots. Remove dots, and match 'city' with ptoassigneend_us_citylist2. SQL Code:

CREATE TABLE ptoassigneend_us_identify4 AS
SELECT *
FROM ptoassigneend_us_temp5
WHERE city_extracted2 IN (
   SELECT citylist
   FROM ptoassigneend_us_citylist2) AND
   state IS NOT NULL AND state !=  AND
   postcode_extracted ~* '\d{5}';
SELECT 38

The remaining records are stored in ptoassigneend_us_temp5.

Summary

Table Name                                          | Records #
----------------------------------------------------|-------------
ptoassigneend_allus                                 | 3572605
----------------------------------------------------|-------------
ptoassigneend_us_identify0                          | 5343
ptoassigneend_us_temp                               | 3567261 
ptoassigneend_us_identify1                          | 2511356
ptoassigneend_us_temp2                              | 1055874
ptoassigneend_us_identify2                          | 14508
ptoassigneend_us_temp3                              | 1041366
ptoassigneend_us_identify3                          | 664524
ptoassigneend_us_temp4                              | 376835
ptoassigneend_us_identify4                          | 38
ptoassigneend_us_temp5                              | 376797
----------------------------------------------------|-------------
ptoassigneend_us_identify_subtotal                  | 3195769
----------------------------------------------------|-------------
ptoassigneend_us_candid1 (city and state are clean)  | 136958 
ptoassigneend_us_candid2 (postcode is clean)        | 184123 

Output: ptoassigneend_us_identify_subtotal

Union ptoassigneend_us_identify(0-4) to generate ptoassigneend_us_identify_subtotal with clean city, state and postcode. This table contains 89.5% of all the records in ptoassigneend_allus. 10.5% left in ptoassigneend_us_temp5.

  Table "public.ptoassigneend_us_identify_subtotal"
     Column      |          Type          | Modifiers
-----------------|------------------------|-----------
reelno           | integer                |
frameno          | integer                |
name             | character varying(500) |
addrline1        | character varying(500) |
addrline2        | character varying(500) |
city_cleaned     | text                   |
state_cleaned    | text                   |
country          | character varying(500) |
postcode_cleaned | text                   |

Output: ptoassigneend_us_candid1

One problem of records in ptoassigneend_us_temp5 is that the postcode is missing.

ptoassigneend_us_candid1 is a subset of ptoassigneend_us_temp5. It contains clean city and state info, but postcode is missing.

SQL code:

CREATE TABLE ptoassigneend_us_candid1 AS
SELECT *
FROM ptoassigneend_us_temp5
WHERE city_extracted2 IN (
   SELECT citylist
   FROM ptoassigneend_us_citylist2) AND
   state IS NOT NULL AND state != ;
SELECT 136958 

Remaining records are in table ptoassigneend_us_temp6 (SELECT 239837).

Output: ptoassigneend_us_candid2

ptoassigneend_us_candid2 is also a subset of ptoassigneend_us_temp5. It contains clean postcode info, but city and state are not identified.

I randomly checked the city_extracted in ptoassigneend_us_candid2, and it is quite clean. Some city records are misspelt, such as 'Oklahama City'. We may identify clean city based on the length of records.

Note: About 60 records are missing. For example, the # of records in ptoassigneend_us_temp + # of records in ptoassigneend_us_identify0 != # ptoassigneend_allus.

To do

  • Remove city, state, zip, country from addrline1 & addrline2 to get clean addrlines.
  • Maybe concatenate addrline1 and addrline to make addrline