Part-of-speech tagging for ranking search results

Patent metadata

Field	Value
Publication number	US-9514221-B2
Application number	US-201313828316-A
Country	US
Kind code	B2
Filing date	Mar 14, 2013
Priority date	Mar 14, 2013
Publication date	Dec 6, 2016
Grant date	Dec 6, 2016

Abstract

Systems, methods, and computer-readable storage media are provided for utilizing part-of-speech (POS) tagging of both the words included in a search query and the words included in potential search result documents to improve query alteration accuracy and search result ranking. Upon receiving a search query, POS tags are assigned to the words comprising the query to create query word-tag pairs. The query word-tag pairs are utilized to reformulate the query and are compared with document word-tag pairs included in a plurality of potential search result documents to determine a degree of similarity. The degree of similarity is utilized as an input in scoring and/or ranking the relevance of the potential search result documents with respect to one another.

First claim

What is claimed is:

1. One or more computer-readable storage media storing computer-useable instructions that, when used by one or more computing devices, cause the one or more computing devices to perform a method for utilizing part-of-speech tagging in ranking potential search result documents, the method comprising:
receiving a search query including one or more words;
reformulating the received query by assigning part-of-speech tags to at least a portion of the one or more words included in the search query to create one or more query word-tag pairs;
identifying document word-tag-pairs in a document word-tag-pairs searchable index using the reformulate query to compare the one or more query word-tag pairs to the one or more document word-tag pairs included in a plurality of potential search result documents to determine a degree of similarity; and
using the degree of similarity to score the relevance of each of the plurality of potential search result documents to rank the plurality of potential search result documents based on the comparison of the query word-tag pairs from the search query and the document word-tag pairs from the plurality of potential search result documents;
wherein upon the part-of-speech tags assigned to the one or more words identifying at least one of the one or more words as a noun, the respective part-of- speech tag further identifies the noun as plural, possessive, or singular; and
wherein upon the part-of-speech tags assigned to the one or more words identifying at least one of the one or more words as a verb, the respective part-of-speech tag further identifies the verb based upon tense or aspect.

2. The one or more computer-readable storage media of claim 1 , wherein the part-of-speech tags assigned to the one or more words identifies the one or more words as one of a noun, a verb, an adjective, an adverb, a determiner, an article, a preposition, a pronoun, a conjunction, and an interjection.

3. The one or more computer-readable storage media of claim 1 , wherein using the degree of similarity to score the relevance of each of the plurality potential search result documents comprises using the degree of similarity for each potential search result document to rank the potential search result documents of the plurality of potential search result documents.

4. The one or more computer-readable storage media of claim 1 , wherein the method further comprises using the part-of-speech tags assigned to the portion of the one or more words included in the search query to reformulate the search query.

5. The one or more computer-readable storage media of claim 1 , wherein at least one potential search result document of the plurality of potential search result documents includes one or more words, wherein the method further comprises assigning part-of-speech tags to at least a portion of the one or more words included in the at least one potential search result document to create the one or more document word-tag pairs.

6. The one or more computer-readable storage media of claim 5 , wherein the method further comprises indexing the document word-tag pairs in a searchable index.

7. The one or more computer-readable storage media of claim 5 , wherein each part-of-speech tag is one of a grammatical part-of-speech tag and a syntactical part-of-speech tag.

8. A method being performed by one or more computing devices including at least one processor, the method for utilizing part-of-speech tagging in ranking potential search result documents, the method comprising:
reformulating a received search query by assigning part-of-speech tags to words in the search query to create query word-tag pairs;
assigning part-of-speech tags to words in a plurality of potential search result documents to create document word-tag pairs searchable index;
identifying the document word-tag-pairs in the searchable index using the reformulate query;
utilizing a degree of similarity score between matching query word-tag pairs and document word-tag pairs to rank the plurality of potential search result documents relative to one another, wherein a first search result document having a matching query word-tag pair and document word-tag pair is ranked higher than a second search result having a word match with the search query where the word match has a different part-of-speech tag in the second search result than in the search query,
wherein upon the part-of-speech tags assigned to the one or more words identifying at least one of the one or more words as a noun, the respective part-of- speech tag further identifies the noun as plural, possessive, or singular; and
wherein upon the part-of-speech tags assigned to the one or more words identifying at least one of the one or more words as a verb, the respective part-of-speech tag further identifies the verb based upon tense or aspect.

9. The method of claim 8 , wherein utilizing matching query word-tag pairs and document word-tag pairs to rank the plurality of potential result documents relative to one another comprises:
comparing the query word-tag pairs with the document word-tag pairs to determine a degree of similarity for each document of the plurality of potential search result documents; and
utilizing the respective degrees of similarity as an input to rank the plurality of potential search result documents relative to one another.

10. The method of claim 8 , wherein the part-of-speech tags assigned to the words identifies the words as nouns, verbs, adjectives, adverbs, determiners, articles, prepositions, pronouns, conjunctions, and interjections.

11. The method of claim 8 , wherein the method further comprises using the part-of-speech tags assigned to the words in the search query to reformulate the search query.

12. The method of claim 8 , wherein the method further comprises indexing the document word-tag pairs in a searchable index.

13. A system comprising:
an information retrieval engine having one or more processors and one or more computer-readable storage media;
a data store coupled with the information retrieval engine, wherein the information retrieval engine:
receives a query including one or more words;
assigns part-of-speech tags to at least a portion of the one or more words included in the query to create query word-tag pairs, wherein a first word of the search query is associated with a first query word-tag pair;
reformulates the query using the assigned part-of-speech tags; and
utilizing the reformulated query, determines potential documents for retrieval by matching the query word-tag pairs of the search query with document word-tag pairs included in the potential documents and determine a degree of similarity to rank the search query result,
wherein the matching includes identifying the first word in a first document and a second document, wherein the first document includes a query word-tag pair and document word-tag pair match for the first word, and wherein the second document has a document word-tag pair for the first word that is different from the query word-tag pair for the first word,
wherein upon the part-of-speech tags assigned to the one or more words identifying at least one of the one or more words as a noun, the respective part-of- speech tag further identifies the noun as plural, possessive, or singular; and
wherein upon the part-of-speech tags assigned to the one or more words identifying at least one of the one or more words as a verb, the respective part-of-speech tag further identifies the verb based upon tense or aspect.

14. The system of claim 13 , wherein the information retrieval engine further determines a degree of similarity betw

Assignees

• Microsoft Technology Licensing Llc

Inventors

• Tropin Kyrylo
• Sia Ka Cheung
• Xu Gu
• Middha Bhuvan
• Yao Qi
• Dey Sandeep
• Tang Junbiao

Classifications

• G06F16/3344 Primary

  using natural language analysis · CPC title

• G06F17/30684 Primary

  Physics · mapped topic