IPC분류정보
| 국가/구분 |
United States(US) Patent
등록
|
| 국제특허분류(IPC7판) |
|
| 출원번호 |
US-0567667
(2009-09-25)
|
| 등록번호 |
US-9009177
(2015-04-14)
|
|
발명자
/ 주소 |
- Zheng, Yu
- Sheng, Chang
- Xie, Xing
|
| 출원인 / 주소 |
|
| 대리인 / 주소 |
|
| 인용정보 |
피인용 횟수 :
16 인용 특허 :
102 |
초록
▼
Techniques for searching and providing geographical regions are described. The process searches and recommends points of interests based on a user-specified region. Points of interests include spatial objects (e.g., buildings, landmarks, rivers, parks) and their distributions in a geographical regio
Techniques for searching and providing geographical regions are described. The process searches and recommends points of interests based on a user-specified region. Points of interests include spatial objects (e.g., buildings, landmarks, rivers, parks) and their distributions in a geographical region. The process searches and recommends points of interests by partitioning a spatial map into grids to identify representative categories located in each of the grids. In response to the user-specified region, a set of geographical candidates containing the representative categories is retrieved. The process determines whether the user-specified region and the set of geographical candidates include similar or common representative categories and similar or common spatial distributions of the representative categories. Then the process provides the top ranked set of geographical candidates that have similar content information.
대표청구항
▼
1. A method implemented at least partially by one or more processors, the method comprising: accessing a spatial map;partitioning the spatial map into grids to identify representative categories located in each of the grids;selecting a query region with a first two or more points of interests in eac
1. A method implemented at least partially by one or more processors, the method comprising: accessing a spatial map;partitioning the spatial map into grids to identify representative categories located in each of the grids;selecting a query region with a first two or more points of interests in each of one or more representative categories in the spatial map;in response to the selecting the query region, receiving a set of geographical region candidates in the spatial map that includes a second two or more points of interests in each of the one or more representative categories;selecting one or more reference points at the spatial map;calculating a spatial distribution of the first two or more points of interests, the spatial distribution including a reference distance that is an average distance of the first two or more points of interests to the one or more reference points;calculating spatial distributions of the second two or more points of interests of candidates of the set of geographical region candidates, the spatial distribution of the second two or more points of interests including a reference distance that is an average distance of the second two or more points of interests to the one or more reference points;calculating respective similarities between the query region and candidates of the set of candidates in the spatial map at least based on respective spatial distributions of the query region and respective candidate; andproviding top ranking geographical regions from the set of geographical region candidates based on the similarities. 2. The method of claim 1, wherein the spatial map comprises at least one of a spatial dataset of locations based on the representative categories identified for published information of businesses, museums, restaurants, sight attractions; a spatial dataset identifying objects as rivers, lakes, parks, and buildings; or a spatial dataset of a geographical region. 3. The method of claim 1, wherein the partitioning comprises using a hierarchical structure with a root node denoting a region and imposing grids on the spatial map to correspond to partitioned cells from a parent's cell. 4. The method of claim 1, further comprising building an inverted-index over grids to facilitate a search for the set of geographical region candidates. 5. The method of claim 1, wherein the partitioning the spatial map into grids identifies the representative categories for the query region and each of the grids to be searched based on a quadtree structure and an inverted tree list. 6. The method of claim 1, wherein the calculating the spatial distribution comprises: estimating a geographical correlation of categories of the query region and a respective geographic region candidate in the set of geographical region candidates; andevaluating a similarity between the query region and the respective geographic region candidate in the set of geographical region candidates using a vector associated with different categories of the points of interests. 7. The method of claim 6, wherein the calculating the spatial distribution further comprises determining a mutual distance, wherein the mutual distance is a distance between at least two geographical region candidates. 8. The method of claim 1, further comprising pruning the set of geographical region candidates based on a category-based pruning and a spatial feature-based pruning. 9. The method of claim 1, wherein the calculating respective similarities between the query region and the set of candidates in the spatial map further comprises having common content information comprising: common geometric properties that include a scale and a shape; andcommon content properties that include common points of interests and common representative categories. 10. A system comprising: a memory;one or more processors coupled to the memory to perform acts comprising:providing a spatial map containing geographical regions partitioned with points of interests based at least in part on representative categories;identifying a user-specified region with two or more points of interests in each of one or more representative categories in the spatial map;searching for a set of geographical region candidates with another two or more points of interests in each of the one or more representative categories based at least in part on a spatial similarity to comparable points of interest in a representative category in the user-specified region, the spatial similarity comprises comparable distribution of the two or more points of interests in the representative categories; andpresenting top geographical region candidates based on a result of the searching. 11. The system of claim 10, wherein the geographical regions are partitioned by imposing a grid on the spatial map to determine an inverse region frequency of the representative category. 12. The system of claim 10, further comprising calculating the spatial similarity, the calculating the spatial similarity comprises using a spatial vector space model to determine at least one of: a mutual distance, wherein the mutual distance is a distance between at least two geographical region candidates; ora reference distance, wherein the reference distance is an average distance of the geographical region candidates to a reference point. 13. The system of claim 10, wherein searching for a set of geographical region candidates with points of interests in representative categories further comprises determining a content similarity comprising points of interests and representative categories. 14. The system of claim 10, further comprising performing representative categories pruning and spatial feature-based pruning on the set of geographical region candidates. 15. One or more processors to perform acts comprising: selecting a query region with a first two or more points of interest in each of one or more representative categories in a spatial map;in response to the query region, receiving a set of geographical region candidates in the spatial map that includes a second two or more points of interests in each of the one or more representative categories;calculating respective similarities between the query region and the region candidates of the set of geographical region candidates in the spatial map at least based on spatial distribution of the two or more points of interests in the query region and spatial distributions of comparable points of interests in the candidates of the set of geographical region candidates, a respective spatial distribution including a respective distribution of respective points of interests in respective representative categories; andpresenting top geographical regions from the set of candidates based on the similarities. 16. The one or more processors of claim 15, the acts further comprising providing a spatial map containing geographical regions partitioned with points of interests based at least in part on the representative categories. 17. The one or more processors of claim 15, wherein the calculating the respective similarities between the query region and the set of candidates in the spatial map based on one or more properties of the query region and the set of candidates further comprises: calculating the respective similarities between the query region and candidate regions of the set of candidates in the spatial map based on respective geometric properties that include a scale and a shape of the query region and the set of geographical region candidates; and respective content properties that include points of interests and representative categories of the query region and the candidate regions of the set of geographical region candidates. 18. The one or more processors of claim 15, the acts further comprising analyzing a shape and a size of the query region to determine a quadtree layer to initiate a similar region search for the candidate regions of the set of geographical region candidates. 19. The one or more processors of claim 15, the acts further comprising performing representative categories pruning and spatial feature-based pruning on the set of geographical region candidates. 20. The one or more processors of claim 15, the acts further comprising identifying a start level in a quadtree and gradually expanding to regions of a shape similar and similar value to the query region.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.