Method and apparatus for preparation of indexing structures for determining similar points-of-interests
View Patent ↗An approach is provided for determining the similarity between two points-of-interest. Data about a plurality of features of each of a plurality of points-of-interest are received. A feature vector is created based on the respective features for each of the points-of-interest. A geospatial index comprising the feature vectors and associated location data for each of the points-of-interest is generated.
1. A method comprising:
receiving data about a plurality of features for each of a plurality of points-of-interest;
creating, based on the respective features, a feature vector for each of the points-of-interest;
generating a geospatial index comprising the feature vectors and associated location data for each of the points-of-interest;
selecting one of the points-of-interest as a reference point-of-interest; and
determining, by a computer processor, a similarity score with respect to the reference point-of-interest for each of the other points-of-interest based on the corresponding feature vectors and the feature vector of the reference point-of-interest.
2. The method of claim 1 , further comprising: receiving training set input specifying feature vectors of similar points-of-interest; and determining a weighting vector for use with the reference point-of-interest based on the training set.
3. The method of claim 2 , wherein the respective similarity scores for each of the other points-of-interest is further based on the weighting vector.
4. The method of claim 3 , wherein the respective similarity scores are determined by computing an inner vector product of the weighting vector, the feature vector of the reference point-of-interest and the corresponding feature vector for each other point-of-interest.
5. The method of claim 1 , wherein the features include classification features, tag features, price features, ratings features, text-based description features, or a combination thereof.
6. The method of claim 1 , wherein the features include text-based description features based on editorial content or user reviews.
7. An apparatus comprising:
at least one processor; and
at least one memory including computer program code,
the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to perform at least the following,
receive data about a plurality of features for each of a plurality of points-of-interest;
create, based on the respective features, a feature vector for each of the points-of-interest;
generate a geospatial index comprising the feature vectors and associated location data for each of the points-of-interest;
select one of the points-of-interest as a reference point-of-interest; and
determine a similarity score with respect to the reference point-of-interest for each of the other points-of-interest based on the corresponding feature vectors and the feature vector of the reference point-of-interest.
8. The apparatus of claim 7 , wherein the apparatus is further caused, at least in part, to:
receive training set input specifying feature vectors of similar points-of-interest; and determine a weighting vector for use with the reference point-of-interest based on the training set.
9. The apparatus of claim 8 , wherein the respective similarity scores for each of the other points-of-interest is further based on the weighting vector.
10. The apparatus of claim 9 , wherein the respective similarity scores are determined by computing an inner vector product of the weighting vector, the feature vector of the reference point-of-interest and the corresponding feature vector for each other point-of-interest.
11. The apparatus of claim 7 , wherein the features include classification features, tag features, price features, ratings features, text-based description features, or a combination thereof.
12. The apparatus of claim 7 , wherein the features include text-based description features based on editorial content or user reviews.
13. A computer-readable storage medium carrying one or more sequences of one or more instructions which, when executed by one or more processors, cause an apparatus to at least perform the following steps:
receiving data about a plurality of features for each of a plurality of points-of-interest;
creating, based on the respective features, a feature vector for each of the points-of-interest;
generating a geospatial index comprising the feature vectors and associated location data for each of the points-of-interest;
selecting one of the points-of-interest as a reference point-of-interest; and
determining a similarity score with respect to the reference point-of-interest for each of the other points-of-interest based on the corresponding feature vectors and the feature vector of the reference point-of-interest.
14. The computer-readable storage medium of claim 13 , wherein the apparatus is caused, at least in part, to further perform:
receiving training set input specifying feature vectors of similar points-of-interest; and determining a weighting vector associated with the reference point-of-interest based on the training set.
15. The computer-readable storage medium of claim 14 , wherein the respective similarity scores for each of the other points-of-interest is further based on the weighting vector.
16. The computer-readable storage medium of claim 15 , wherein the respective similarity scores are determined by computing an inner vector product of the weighting vector, the feature vector of the reference point-of-interest and the corresponding feature vector for each other point-of-interest.
17. The computer-readable storage medium of claim 13 , wherein the features include classification features, tag features, price features, ratings features, text-based description features, or a combination thereof.