Methods for automatic generation of parallel corpora

• US 9,881,006 B2
• Filed: 12/31/2014
• Issued: 01/30/2018
• Est. Priority Date: 02/28/2014
• Status: Active Grant

• Current Assignee: PayPal, Inc. (PayPal Holdings, Inc.)
• Original Assignee: PayPal, Inc. (PayPal Holdings, Inc.)
• Inventors: Ruvini, Jean-David, Sawaf, Hassan, Barnes, Derek
• Primary Examiner(s): Zhu, Richard
• Application Number:
  US14/587,552
• Publication Number:
  US 20150248401A1
• Time in Patent Office: 1,126 Days
• CPC Class Codes: G06F 40/40 Processing or translation of natural language natural language analysis G06F40/20; semantic analysis G06F40/30
• CPC Class Codes: G06F 40/42 Data-driven translation
• CPC Class Codes: G06F 40/44 Statistical methods, e.g. probability models
• CPC Class Codes: G06F 40/45 Example-based machine translation; Alignment
• CPC Class Codes: G06F 40/49 using very large corpora, e.g. the web
• CPC Class Codes: G06F 40/58 Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation
• CPC Class Codes: G06Q 30/0601 Electronic shopping [e-shopping]

Abstract

A method of forming parallel corpora comprises receiving sets of items in first language and second languages, each of the sets having one or more associated descriptions and metadata. The metadata is collected from the two sets of items and are aligned using the metadata. The aligned metadata are mapped from the first language to the second language for each of the sets. The descriptions of two items are fetched and the structural similarity of the descriptions is measured to assess whether two items are likely to be translations of each other. For mapped items with structurally similar descriptions, the mapped item descriptions are formed into respective sentences in first language and in the second language. The sentences are parallel corpora which may be used to translate an item from the first language to the second language, and also to train a machine translation system.

Claims

1. A computer implemented method comprising:

receiving a first set of item listings for the sale of products or services in a first language and a second set of item listings for the sale of products or services in a second language, each of the item listings in the first and second sets of item listings comprising one or more descriptions and metadata identifying the products or services corresponding to the respective item listing;

collecting the metadata from the first and second sets of item listings and aligning, using the collected metadata identifying the products or services, a first item listing of the first set of item listings with a second item listing of the second set of item listings in which the first item listing and the second item listing are aligned based on the first item listing and the second item listing being directed toward the same products or services;

mapping the first item listing to the second item listing based on the aligning of the first item listing with the second item listing;

fetching a first description of the first item listing and a second description of the second item listing;

measuring the structural similarity of the fetched first description with respect to the fetched second description to assess whether the first description and the second description are likely to be translations of each other; and

in response to the first description and the second description being structurally similar, forming the first description into a first sentence in the first language as a translation of the second description into the first language and forming the second description into a second sentence in the second language as a translation of the first description into the second language.

2. The method of claim 1 wherein the first sentence and the second sentence comprise parallel corpora, the method further comprising using the parallel corpora to translate an item listing from the first language to the second language.

3. The method of claim 1, further comprising sorting the metadata based on one of a bar code, an item identifier, an item title, a Universal Product Code (UPC) aspect name, a European Article Number (EAN) aspect name, a UPC aspect value, an EAN aspect value, an International Standard Book Number (ISBN), a category identifier, a site identifier, a seller identifier, a picture of the item, or an auction end date.

4. The method of claim 1 wherein the metadata is collected by one of an API to a web-based metadata collection module or a by stand-alone metadata collection module.

5. The method of claim 1 wherein measuring the structural similarity of the first item listing and the second item listing comprises using metadata to align the item titles and the item descriptions that correspond to the first item listing and the second item listing.

6. The method of claim 4 wherein measuring the structural similarity of the first item listing and the second item listing is based on a hypertext markup language (HTML) code of the item descriptions.

7. The method of claim 1 wherein the first sentence and the second sentence comprise parallel corpora, the method further comprising using the parallel corpora to train a machine translation system.

8. One or more computer-readable hardware storage device having embedded therein a set of instructions which, in response to being executed by one or more processors of a system, causes the system to execute operations comprising:

receiving a first set of item listings for the sale of products or services in a first language and a second set of item listings for the sale of products or services in a second language, each of the item listings in the first and second sets of item listings comprising one or more descriptions and metadata identifying the products or services corresponding to the respective item listing;

collecting the metadata from the first and second sets of item listings and aligning, using the collected metadata identifying the products or services, a first item listing of the first set of item listings with a second item listing of the second set of item listings in which the first item listing and the second item listing are aligned based on the first item listing and the second item listing being directed toward the same products or services;

mapping the first item listing to the second item listing based on the aligning of the first item listing with the second item listing;

fetching a first description of the first item listing and a second description of the second item listing;

measuring the structural similarity of the fetched first description with respect to the fetched second description to assess whether the first description and the second description are likely to be translations of each other; and

in response to the first description and the second description being structurally similar, forming the first description into a first sentence in the first language as a translation of the second description into the first language and forming the second description into a second sentence in the second language as a translation of the first description into the second language.

9. The one or more computer-readable hardware storage device of claim 8 wherein the first sentence and the second sentence comprise parallel corpora, the operations further comprising using the parallel corpora to translate an item listing from the first language to the second language.

10. The one or more computer-readable hardware storage device of claim 8, wherein the operations further comprise sorting the metadata based on one of a bar code, an item identifier, an item title, a Universal Product Code (UPC) aspect name, a European Article Number (EAN) aspect name, a UPC aspect value, an EAN aspect value, an International Standard Book Number (ISBN), a category identifier, a site identifier, a seller identifier, a picture of the item, or an auction end date.

11. The one or more computer-readable hardware storage device of claim 8 wherein the metadata is collected by one of an API to a web-based metadata collection module or a by stand-alone metadata collection module.

12. The one or more computer-readable hardware storage device of claim 8 wherein measuring the structural similarity of the first item listing and the second item listing comprises using metadata to align the item titles and the item descriptions that correspond to the first item listing and the second item listing.

13. The one or more computer-readable hardware storage device of claim 11 wherein measuring the structural similarity of the first item listing and the second item listing is based on a hypertext markup language (HTML) code of the item descriptions.

14. The one or more computer-readable hardware storage device of claim 8 wherein the first sentence and the second sentence comprise parallel corpora, the operations further comprising using the parallel corpora to train a machine translation system.

15. A system comprising:

one or more computer-readable hardware storage devices having embedded therein a set of instructions; and

one or more hardware processors communicatively coupled to the one or more computer-readable hardware storage devices, and configured to, in response to execution of the set of instructions, cause the system to perform operations, the operations comprising;

receiving a first set of item listings for the sale of products or services in a first language and a second set of item listings for the sale of products or services in a second language, each of the item listings in the first and second sets of item listings comprising one or more descriptions and metadata identifying the products or services corresponding to the respective item listing;

collecting the metadata from the first and second sets of item listings and aligning, using the collected metadata identifying the products or services, a first item listing of the first set of item listings with a second item listing of the second set of item listings in which the first item listing and the second item listing are aligned based on the first item listing and the second item listing being directed toward the same products or services;

mapping the first item listing to the second item listing based on the aligning of the first item listing with the second item listing;

fetching a first description of the first item listing and a second description of the second item listing

measuring the structural similarity of the fetched first description with respect to the fetched second description to assess whether the first description and the second description item listings are likely to be translations of each other; and

in response to the first description and the second description being structurally similar, forming the first description into a first sentence in the first language as a translation of the second description into the first language and forming the second description into a second sentence in the second language as a translation of the first description into the second language.

16. The one or more hardware processors of claim 15 wherein the first sentence and the second sentence comprise parallel corpora, the operations further comprising using the parallel corpora to translate an item listing from the first language to the second language.

17. The one or more hardware processors of claim 15 wherein the operations further comprise sorting the metadata based on one of a bar code, an item identifier, an item title, a Universal Product Code (UPC) aspect name, a European Article Number (EAN) aspect name, a UPC aspect value, an EAN aspect value, an International Standard Book Number (ISBN), a category identifier, a site identifier, a seller identifier, a picture of the item, or an auction end date.

18. The one or more hardware processors of claim 15 wherein the metadata is collected by one of an API to a web-based metadata collection module or a by stand-alone metadata collection module.

19. The one or more hardware processors of claim 15 wherein measuring the structural similarity of the first item listing and the second item listing comprises using metadata to align the item titles and the item descriptions that correspond to the first item listing and the second item listing.

20. The one or more hardware processors of claim 18 wherein measuring the structural similarity of the first item listing and the second item listing is based on a hypertext markup language (HTML) code of the item descriptions.