User retrieval enhancement

• US 10,235,470 B2
• Filed: 12/05/2014
• Issued: 03/19/2019
• Est. Priority Date: 12/06/2013
• Status: Active Grant

• Current Assignee: HERE Global B.V. (HERE Holding Corporation)
• Original Assignee: HERE Global B.V. (HERE Holding Corporation)
• Inventors: Gur, Nadav, Young, Stephen, Melnik, Ofer, Goldberg, Imri
• Primary Examiner(s): Mamillapalli, Pavan
• Application Number:
  US14/562,334
• Publication Number:
  US 20150161271A1
• Time in Patent Office: 1,565 Days
• Field of Search: 707728
• CPC Class Codes: G06F 16/9537 Spatial or temporal dependent retrieval, e.g. spatiotemporal queries

Abstract

Systems and methods are described for performing a travel search and presenting search results to a user. A computing device may retrieve multiple travel documents and analyze the travel documents to identify points of interest. The computing device may analyze the subset of the travel documents that have points of interest or significant information regarding points of interest. The computing device is configured to calculate a POI relevance score based on text associated with the identified POIs. The POI relevance score indicates how likely the text is relevant to a travel search and rank the subset of the travel documents based on the POI relevance score.

Claims

1. A method of improving user retrieval of travel search results, the method comprising:

retrieving, using a processor, a plurality of documents in response to a travel search query;

identifying, using the processor, a subset of travel documents within the plurality of documents based on text of the plurality of documents, wherein the travel documents contain data indicative of travel;

sorting, by the processor, the subset of travel documents into one or more indexes based on a type of travel search query;

selecting, by the processor, an index of the one or more indexes based on the type of travel search query;

identifying, using the processor, points of interest (POIs) referenced by or described in the travel documents of the selected index based on text of the travel documents of the selected index, wherein the POIs are associated with and relevant to the type of travel search query, and wherein identifying the POIs comprises tagging, by the processor, the travel documents containing the POIs;

identifying, using the processor, travel traits for the identified POIs based on text of the travel documents of the selected index, wherein the travel traits comprise facts that describe the POIs, the facts being indicative of travel;

calculating, using the processor, a POI relevance score for each travel document of the selected index based on text contained within each travel document of the selected index associated with the identified POIs, wherein the POI relevance score indicates how likely the text is relevant to the travel search query;

calculating, using the processor, a trait inclusion score for each travel document of the selected index based on the identified travel traits associated with the identified POIs, wherein the trait inclusion score indicates the type or level of detail of description of the POI in the travel documents of the selected index; and

providing the travel documents of the selected index in a ranked order based on the POI relevance score and the trait inclusion score.

2. The method of claim 1, further comprising:

analyzing text contained in the travel documents of the selected index associated with at least one of the points of interest to determine an ambiguity score;

comparing the ambiguity score to a threshold; and

removing one or more of the travel documents of the selected index in response to the ambiguity score being above the threshold.

3. The method of claim 1, further comprising:

storing a ranked subset of the travel documents of the selected index for each of the POIs.

4. The method of claim 1, further comprising:

identifying a type of travel document from text of the travel documents of the selected index; and

assigning a travel type score based on the type of travel document.

5. The method of claim 1, further comprising:

extracting a portion of one of the travel documents of the selected index based on a respective POI relevance score.

6. The method of claim 1, further comprising:

identifying, using the processor, travel amenities for the POIs from the travel documents of the selected index.

7. The method of claim 1, further comprising:

identifying, using the processor, whether one of the POIs dominates a respective one of the travel documents of the selected index.

8. The method of claim 1, further comprising:

identifying, using the processor, reviews for the POIs from the travel documents of the selected index.

9. The method of claim 1, further comprising:

receiving a travel search from a user;

identifying a POI and a least one qualifying query from the travel search;

accessing the travel documents ranked based on the POI relevance and trait inclusion scores; and

providing the travel documents in an order based on the ranking of the travel documents from the POI relevance and trait inclusion scores and based on the at least one qualifying query.

10. The method of claim 1, wherein the trait inclusion score is based on a number of traits, types of traits, or combinations thereof.

11. An apparatus for improving user retrieval of travel search results, the apparatus comprising:

at least one processor; and

at least one memory including computer program code for one or more programs;

the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to at least perform;

identify, in response to a travel search, a plurality of travel documents, wherein the plurality of travel documents contain data indicative of travel;

sort the plurality of travel documents into one or more indexes based on a type of travel search;

select an index of the one or more indexes based on the type of travel search;

identify points of interest (POIs) contained in the travel documents of the selected index, wherein identifying the POIs comprises tagging, by the processor, the travel documents containing the POIs;

identify travel traits for the identified POIs within the travel documents of the selected index, wherein the travel traits comprise facts that describe the POIs, the facts being indicative of travel;

calculate a POI relevance score for each travel document in the selected index based on text associated with the identified POIs, wherein the POI relevance score indicates how likely the text is relevant to the travel search;

calculate a trait inclusion score for each travel document in the selected index based on the identified travel traits associated with the identified POIs, wherein the trait inclusion score indicates the type or level of detail of description of the POIs in the travel document; and

store respective POI relevance and trait inclusion scores for the travel documents of the selected index.

12. The apparatus of claim 11, wherein the at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus to at least perform:

receive a travel search including data indicative of a POI; and

generate search results based on the POI relevance and trait inclusion scores.

13. The apparatus of claim 12, wherein the travel search includes data indicative of a qualifier for the POI, and the at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus to at least perform:

filter the search results according to the qualifier.

14. The apparatus of claim 11, wherein the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to at least perform:

determine ambiguity scores for the points of interest based on text within the travel documents of the selected index associated with the points of interest;

compare the ambiguity scores to a threshold; and

remove one or more of the travel documents of the selected index in response to one or more of the ambiguity scores being above the threshold.

15. The apparatus of claim 11, wherein the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to at least perform:

store a ranked subset of the travel documents of the selected index for each of the POIs.

16. The apparatus of claim 11, wherein the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to at least perform:

identify a type of travel document from text of the travel documents of the selected index; and

assign a travel type score based on the type of travel document.

17. The apparatus of claim 11, wherein the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to at least perform:

extract a portion of one of the travel documents of the selected index based on a respective POI relevance or trait inclusion score.

18. A method of improving user retrieval of travel search results, the method comprising:

receiving a travel search from a user;

identifying, using a processor, a point of interest (POI) and a least one qualifying query from the travel search;

identifying, using the processor based on the travel search, a plurality of documents in response to the travel search;

sorting, by the processor, the plurality of documents into one or more indexes based on a type of travel search;

selecting, by the processor, an index of the one or more indexes based on the type of travel search;

calculating, using the processor, a POI relevance score and a trait inclusion score for each document of the selected index, wherein the POI relevance score indicates how likely each document of the selected index is relevant to the travel search and wherein the trait inclusion score indicates how well travel related details are known about the POI in each document of the selected index;

receiving a subset of the documents of the selected index ranked based at least on the POI relevance score and the trait inclusion score; and

displaying the subset of the documents of the selected index in an order as ranked from the POI relevance and trait inclusion scores, by the processor, and based on the at least one qualifying query.

19. The method of claim 18, further comprising:

calculating a matching score for each of the documents of the subset of the documents of the selected index based on the at least one qualifying query.

20. An apparatus comprising:

at least one processor; and

at least one memory including computer program code for one or more programs;

the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to at least perform;

identify a travel search from a user, wherein the travel search includes a qualifying term and a point of interest (POI) or a location for the POI;

identify a plurality of documents in response to the travel search;

sort the plurality of travel documents into one or more indexes based on a type of travel search;

select an index of the one or more indexes based on the type of travel search;

calculate a POI relevance score and a trait inclusion score for each document of the selected index, wherein the POI relevance score indicates how likely each document of the selected index is relevant to the travel search and wherein the trait inclusion score indicates how well travel related details are known about the POI in each document of the selected index;

receive a subset of the documents of the selected index ranked based on the POI relevance score and the trait inclusion score; and

display the subset of the documents of the selected index in an order based on a ranking of the documents of the selected index from the POI relevance and trait inclusion scores and based on the at least one qualifying query.