Identification of inaccurate addresses for package deliveries

Patent metadata

Field	Value
Publication number	US-2019005439-A1
Application number	US-201715637231-A
Country	US
Kind code	A1
Filing date	Jun 29, 2017
Priority date	Jun 29, 2017
Publication date	Jan 3, 2019
Grant date	—

Abstract

Systems, methods, and computer-readable media are disclosed for identification of inaccurate addresses for package deliveries. In one embodiment, an example method may include determining a shipping address for a package that includes a product. A recurrent neural network may generate an address deliverability score by providing the shipping address as input. A set of corrective actions corresponding to the address deliverability score may be determined. The address deliverability score and the set of corrective actions may be presented. A selection of a corrective action from the set of corrective actions to implement prior to delivery of the package may be received.

First claim

What is claimed is:

1 . A method comprising:
receiving, by one or more computer processors coupled to at least one memory, from a server, a message comprising information provided by a user device indicative of an online purchase transaction of a product; determining, from the message, a shipping address for the product; generating an address string by concatenating one or more lines of the shipping address from the message; parsing the address string into a set of tokens from the address string, wherein each token of the set of tokens is indicative of a portion of the address string; converting the set of tokens to a plurality of vectors using a word dictionary, wherein text of a token of the set of tokens is replaced with a respective set of integers in the word dictionary; generating, by a recurrent neural network, an address deliverability score, wherein the plurality of vectors are used as input for the recurrent neural network; identifying a set of corrective actions based on the address deliverability score, wherein the set of corrective actions is associated with a score range; presenting the deliverability score and the set of corrective actions; receiving an indication of a selection of a corrective action from the set of corrective actions; and generating a label to affix to the package for the product, wherein the label indicates the corrective action to implement prior to delivery of the package.

2 . The method of claim 1 , wherein the set of corrective actions comprises preventing completion of the online purchase transaction and blocking a cash on delivery payment option.

3 . The method of claim 1 , wherein converting the set of tokens to the plurality of vectors comprises:
identifying at least two word dictionaries by comparing text of the shipping address to words in the at least two word dictionaries; ranking the at least two word dictionaries based on a number of words of the shipping address in each of the two word dictionaries; selecting the word dictionary from the at least two word dictionaries based on the ranking; and determining the respective integer for each token of the set of tokens based on the word dictionary.

4 . The method of claim 1 , wherein generating the address string comprise:
converting text of the address string to lowercase characters.

5 . A method comprising:
determining, by one or more computer processors coupled to at least one memory, a shipping address for a package comprising a product; generating, by a recurrent neural network, an address deliverability score by providing the shipping address as input; determining a set of corrective actions corresponding to the address deliverability score; presenting the address deliverability score and the set of corrective actions; and receiving a selection of a corrective action from the set of corrective actions to implement prior to delivery of the package.

6 . The method of claim 5 , wherein determining the shipping address for the package comprising the product comprises:
determining the shipping address from a message comprising information provided by a user device indicative of an online purchase of the product; or retrieving the shipping address from a remote location.

7 . The method of claim 5 , wherein generating the address deliverability score further comprises:
generating an address string based on the shipping address; generating a set of tokens by parsing the address string; converting the set of tokens to a plurality of vectors using a word dictionary; and generating, by the recurrent neural network, the address deliverability score using the plurality of vectors.

8 . The method of claim 7 , wherein generating the set of tokens comprises:
identifying the word dictionary based on one or more words of the address string; and parsing the address string into the set of tokens using the word dictionary, wherein each token of the set of tokens is indicative of a portion of the shipping address.

9 . The method of claim 7 , wherein converting the set of tokens to the plurality of vectors comprises:
determining a respective integer for each token of the set of tokens based on the word dictionary, wherein a text of a token of the set of tokens is replaced with a respective integer indicated in the word dictionary; and generating the plurality of vectors based on the set of tokens, wherein each vector of the plurality of vectors corresponds to a respective token of the set of tokens.

10 . The method of claim 7 , wherein the shipping address comprises a first line of text and a second line of text and generating the address string based on the shipping address comprises:
generating the address string by concatenating the first line of text and the second line of text and converting the text of the first line and the text of the second line to lowercase characters.

11 . The method of claim 5 , further comprising:
receiving, an indication to generate a label to affix to the package; and generating the label to affix to the package.

12 . The method of claim 11 , wherein the label indicates the corrective action to implement prior to delivery of the package.

13 . The method of claim 5 , wherein the set of corrective actions comprises at least one of assigning delivery of the package to a specific delivery person based on the address deliverability score, requesting additional information for the shipping address from a user, generating a suggested replacement shipping address and presenting the suggested replacement shipping address to the user for confirmation, or flagging the package for a follow-up action.

14 . A device comprising:
at least one memory that stores computer-executable instructions; at least one processor configured to access the at least one memory and execute the computer-executable instructions to:
determine a shipping address for a package comprising a product;
generate, by a recurrent neural network, an address deliverability score by providing the shipping address as input;
determine a set of corrective actions corresponding to the address deliverability score;
present the address deliverability score and the set of corrective actions; and
receive a selection of a corrective action from the set of corrective actions to implement prior to delivery of the package.

15 . The device of claim 14 , wherein, to determine the shipping address for the package comprising the product, the at least one processor is further configured to access the at least one memory and execute the computer-executable instructions to:
determine the shipping address from a message comprising information provided by a user device indicative of an online purchase of the product; or retrieve the shipping address from a remote location.

16 . The device of claim 14 , wherein, to generate the address deliverability score, the at least one processor is further configured to access the at least one memory and execute the computer-executable instructions to:
generate an address string based on the shipping address; generate a set of tokens by parsing the address string; convert the set of tokens to a plurality of vectors using a word dictionary; and generate, by the recurrent neural network, the address deliverability score using the plurality of vectors.

17 . The device of claim 16 , wherein, to generate the set of tokens, the at least one processor is further configured to access the at least one memory and execute the computer-executable instructions to:
identify the word dictionary based on on

Assignees

• Amazon Tech Inc

Inventors

• Nandury Venkata Kishore

Classifications

• G06Q20/00

  Payment architectures, schemes or protocols (apparatus for performing or posting payment transactions G07F7/08, G07F19/00; electronic cash registers G07G1/12) · CPC title

• G06Q30/0635

  replenishment orders; recurring orders · CPC title

• H04L69/40

  for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection (management of faults, events, alarms or notifications in data switching networks H04L41/06) · CPC title

• G06F15/177

  Initialisation or configuration control {(processor initialisation G06F9/4405)} · CPC title

• G06F15/00

  Digital computers in general (details G06F1/00 – G06F13/00); Data processing equipment in general · CPC title