CA2821895A1 - Improving local search ranking using typhoon model - Google Patents
Improving local search ranking using typhoon model Download PDFInfo
- Publication number
- CA2821895A1 CA2821895A1 CA2821895A CA2821895A CA2821895A1 CA 2821895 A1 CA2821895 A1 CA 2821895A1 CA 2821895 A CA2821895 A CA 2821895A CA 2821895 A CA2821895 A CA 2821895A CA 2821895 A1 CA2821895 A1 CA 2821895A1
- Authority
- CA
- Canada
- Prior art keywords
- eye
- size
- search
- band
- weight
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Abandoned
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
A system and method for ranking local search items may include dividing the search area into 2 parts: the eye and the band. The location weight in the item's retrieval score may be a constant within the eye and a decreasing function of distance within the band. The size of the eye may be impacted by query's geo-sensitivity, user's transportation tools and search centre's population density.
Description
IMPROVING LOCAL SEARCH RANKING USING TYPHOON MODEL
FIELD OF THE INVENTION
The present invention relates generally to techniques for information retrieval. More specifically, it relates methods and systems for improving local search performance.
BACKGROUND OF THE INVENTION
Local searching intends to return location associated items within a geographical area.
Location associated items include business lists stored in well-structured databases or websites/web pages with geographical attributes distributed on the World Wide Web.
Geographical area could be a named geographical area such as a city or an artificial search area such as a 100 km extent from user's location. If there is a large amount of items that match the topics in user's query within a geographic area, how to rank search results become a key to improve local search performance.
Distance is the first choice to rank local search results. There are a lot of websites to rank business lists based on distance from user's location. In local web searching, Ryu's invention (US Patent No. 6377961) presented a method to rank search results in an order of relative distance from a user to websites; Ge's invention (US Patent No.
7606798) presented a method to calculate distance scores in ranking search results.
Ranking local search results based on proximity or distance doesn't always provide the best results that user wants. For example, people don't mind going a few kilometres more to have a dinner in a better restaurant. If the restaurant is not in the first 5 or 10 returned results, people may not pay attention to it. O'Clair, Egnor and Greenfield (US
Patent No. 8046371) presented a method to rank local search results based on location prominence factors such as authoritative for a business and total number of documents reference to a business. Qian, Luk and An (US Patent No. 8122013) presented a method to rank business listings based on listing's web popularity.
Without question, distance and other factors that contribute to item's importance should be taken into account in ranking local search results. Based on the concept that people can go a few kilometres more to find a better restaurant, distinguishing their impact geospatially is necessary. Within a certain distance to user's location, distance measure doesn't impact ranking and other factors play main role in ranking.
Outside the area, distance measure play main role in ranking. Therefore, distance's impact to local search results is same as distance's impact to wind and rain in a Typhoon Model: in the center of Typhoon, there is an eye with diameter ranging from 10 to 100 km.
Within the eye is an area of neither fierce wind nor storming. Outside the eye, there is a rain band with storm radius ranging from 100 km to 300 or 400 km. The farther off the center, the weaker the wind and rain is.
SUMMARY OF THE INVENTION
According to one aspect, a method may include dividing the search area into two parts:
the eye and the band for local search, assigning a constant location weight to retrieval score for websites or businesses within the eye, calculating the location weight based on distance for websites or businesses within the band, and ranking the websites within the search area based on the combined weights from topic relevance, website's or business's importance and location.
According to another aspect, a method may include determining the size of the eye based on query's geo-sensitivity, user's transportation tools and search centre's population density. The higher the topic's geo-sensitivity in the query, the smaller the size of the eye is. The faster the transportation tools that client use, the larger the size of the eye is. The higher the population density at the search centre, the smaller the size of the eye is.
According to yet another aspect, a method may include determining the size of the band based on the size of eye. The larger the eye, the larger the size of the band will be for local search. Therefore the size of band is impacted by query's geo-sensitivity, user's transportation tools and location's population density indirectly.
According to a further aspect, a system may include a memory to store instructions and a processor to execute instructions to divide the search area into two parts:
the eye and the band for local search, assign a constant location weight to retrieval score for websites or business list within the eye, calculate the location weight based on distance
FIELD OF THE INVENTION
The present invention relates generally to techniques for information retrieval. More specifically, it relates methods and systems for improving local search performance.
BACKGROUND OF THE INVENTION
Local searching intends to return location associated items within a geographical area.
Location associated items include business lists stored in well-structured databases or websites/web pages with geographical attributes distributed on the World Wide Web.
Geographical area could be a named geographical area such as a city or an artificial search area such as a 100 km extent from user's location. If there is a large amount of items that match the topics in user's query within a geographic area, how to rank search results become a key to improve local search performance.
Distance is the first choice to rank local search results. There are a lot of websites to rank business lists based on distance from user's location. In local web searching, Ryu's invention (US Patent No. 6377961) presented a method to rank search results in an order of relative distance from a user to websites; Ge's invention (US Patent No.
7606798) presented a method to calculate distance scores in ranking search results.
Ranking local search results based on proximity or distance doesn't always provide the best results that user wants. For example, people don't mind going a few kilometres more to have a dinner in a better restaurant. If the restaurant is not in the first 5 or 10 returned results, people may not pay attention to it. O'Clair, Egnor and Greenfield (US
Patent No. 8046371) presented a method to rank local search results based on location prominence factors such as authoritative for a business and total number of documents reference to a business. Qian, Luk and An (US Patent No. 8122013) presented a method to rank business listings based on listing's web popularity.
Without question, distance and other factors that contribute to item's importance should be taken into account in ranking local search results. Based on the concept that people can go a few kilometres more to find a better restaurant, distinguishing their impact geospatially is necessary. Within a certain distance to user's location, distance measure doesn't impact ranking and other factors play main role in ranking.
Outside the area, distance measure play main role in ranking. Therefore, distance's impact to local search results is same as distance's impact to wind and rain in a Typhoon Model: in the center of Typhoon, there is an eye with diameter ranging from 10 to 100 km.
Within the eye is an area of neither fierce wind nor storming. Outside the eye, there is a rain band with storm radius ranging from 100 km to 300 or 400 km. The farther off the center, the weaker the wind and rain is.
SUMMARY OF THE INVENTION
According to one aspect, a method may include dividing the search area into two parts:
the eye and the band for local search, assigning a constant location weight to retrieval score for websites or businesses within the eye, calculating the location weight based on distance for websites or businesses within the band, and ranking the websites within the search area based on the combined weights from topic relevance, website's or business's importance and location.
According to another aspect, a method may include determining the size of the eye based on query's geo-sensitivity, user's transportation tools and search centre's population density. The higher the topic's geo-sensitivity in the query, the smaller the size of the eye is. The faster the transportation tools that client use, the larger the size of the eye is. The higher the population density at the search centre, the smaller the size of the eye is.
According to yet another aspect, a method may include determining the size of the band based on the size of eye. The larger the eye, the larger the size of the band will be for local search. Therefore the size of band is impacted by query's geo-sensitivity, user's transportation tools and location's population density indirectly.
According to a further aspect, a system may include a memory to store instructions and a processor to execute instructions to divide the search area into two parts:
the eye and the band for local search, assign a constant location weight to retrieval score for websites or business list within the eye, calculate the location weight based on distance
2 for websites or business list within the band, and rank the websites or business list within the search area based on the combined weights from topic relevance, website's or business's importance and location.
BRIEF DESCRIPTION OF THE DRAWINGS
Fig. 1 is the exemplary diagram illustrating the concept of Typhoon Model for local search Fig. 2 is the exemplary diagram showing how the size of the eye is impacted by topic's geo-sensitivity Fig. 3 is the exemplary diagram showing how the size of the eye is impacted by user's transportation tools Fig. 4 is the exemplary diagram showing how the size of the eye is impacted by local population density Fig. 5 is a block diagram illustrating an exemplary environment in which embodiments of the present invention may operate Fig. 6 is a flow diagram of an exemplary process for using Typhoon Model to rank websites or business list Fig. 7 is the exemplary diagram showing comparison of results for searching local restaurants ranked by Typhoon Model and distance measurement DETAILED DESCRIPTION
The following detailed description describes the specific embodiments of the invention, examples of which are illustrated in the accompanying drawings. Anyone of ordinary skill in the technology will appreciate that many variations and alterations to the following details are within the scope of the invention.
Fig. 1 is an exemplary diagram illustrating the concept of Typhoon Model for local search.
The local search area is divided into two parts: the eye 101 and the band 102.
There is a centre for the local search area, which is the user's location or a point within
BRIEF DESCRIPTION OF THE DRAWINGS
Fig. 1 is the exemplary diagram illustrating the concept of Typhoon Model for local search Fig. 2 is the exemplary diagram showing how the size of the eye is impacted by topic's geo-sensitivity Fig. 3 is the exemplary diagram showing how the size of the eye is impacted by user's transportation tools Fig. 4 is the exemplary diagram showing how the size of the eye is impacted by local population density Fig. 5 is a block diagram illustrating an exemplary environment in which embodiments of the present invention may operate Fig. 6 is a flow diagram of an exemplary process for using Typhoon Model to rank websites or business list Fig. 7 is the exemplary diagram showing comparison of results for searching local restaurants ranked by Typhoon Model and distance measurement DETAILED DESCRIPTION
The following detailed description describes the specific embodiments of the invention, examples of which are illustrated in the accompanying drawings. Anyone of ordinary skill in the technology will appreciate that many variations and alterations to the following details are within the scope of the invention.
Fig. 1 is an exemplary diagram illustrating the concept of Typhoon Model for local search.
The local search area is divided into two parts: the eye 101 and the band 102.
There is a centre for the local search area, which is the user's location or a point within
3 geographical area that user is interested in. The boundaries of the eye and the search area are circles having the same aforementioned centre. The eye 101 is the inner area enclosed by inner circle, and the band 102 is the area between the inner circle and outer circle. The retrieval score for ranking can be represented as follows:
WT WI + C within the eye S = {WT + W1 + F(d) within the band Where S is the item's retrieval score; WT is the weight for topic relevance;
WI is the weight for item's importance; C is a constant; F(d) is the function of distance. Within the eye for local search, location's weight in the retrieval score is represented by a constant.
Within the band, location's weight in the retrieval score is a decreasing function of distance between the search centre and item's location. The farther off the center, the smaller the location's weight is. The relationship between location weight within the eye and within the band is as follows:
C Max(F(d)) That means that the location weight within the eye is greater than or equal to the highest location weight within the band. The centre of local search area can move in two ways. One way is automatic move when the client uses a device with GPS like smart phone or tablet; another way is that user move the centre to find the interested point in a large area.
Fig. 2 is the exemplary diagram showing how the size of the eye is impacted by topic's geo-sensitivity. Topic's geo-sensitivity refers to the extent of people's decision impacted by changes of geographic location for a topic. The higher a topic's geo-sensitivity, the smaller the size of the eye is. For example, dentist has higher geo-sensitivity than lawyer, so the size of the eye for dentist 201 may be 5 km in radius, and the size of the eye for lawyer 202 may be 50 km in radius. The larger the eye is, the more general the search become. When the size of the eye is more than 40,000 km in radius, the search will become a global search. For example, if you search topic like "java language", it will become a global search. The smaller the eye is, the more local the search become. When
WT WI + C within the eye S = {WT + W1 + F(d) within the band Where S is the item's retrieval score; WT is the weight for topic relevance;
WI is the weight for item's importance; C is a constant; F(d) is the function of distance. Within the eye for local search, location's weight in the retrieval score is represented by a constant.
Within the band, location's weight in the retrieval score is a decreasing function of distance between the search centre and item's location. The farther off the center, the smaller the location's weight is. The relationship between location weight within the eye and within the band is as follows:
C Max(F(d)) That means that the location weight within the eye is greater than or equal to the highest location weight within the band. The centre of local search area can move in two ways. One way is automatic move when the client uses a device with GPS like smart phone or tablet; another way is that user move the centre to find the interested point in a large area.
Fig. 2 is the exemplary diagram showing how the size of the eye is impacted by topic's geo-sensitivity. Topic's geo-sensitivity refers to the extent of people's decision impacted by changes of geographic location for a topic. The higher a topic's geo-sensitivity, the smaller the size of the eye is. For example, dentist has higher geo-sensitivity than lawyer, so the size of the eye for dentist 201 may be 5 km in radius, and the size of the eye for lawyer 202 may be 50 km in radius. The larger the eye is, the more general the search become. When the size of the eye is more than 40,000 km in radius, the search will become a global search. For example, if you search topic like "java language", it will become a global search. The smaller the eye is, the more local the search become. When
4 the size of the eye is 0 km in radius, the search will become pure distance depended local search.
Fig. 3 is the exemplary diagram showing how the size of the eye is impacted by user's transportation tools. When user walks 301, the size of the eye can be just a few hundred meters in radius for searching "restaurant". When user rides a bicycle 302, the size of the eye can be one or two kilometers in radius for searching "restaurant".
When user drives a car 303, the size of the eye can be ten kilometers or more in radius for searching "restaurant". The faster the transportation tool, the larger the size of the eye is.
Fig. 4 is the exemplary diagram showing how the size of the eye is impacted by local population density. In downtown area 401 of Toronto with population density of 10000 people per square kilometer, the eye's size for searching a coffee shop may be a few hundreds of meters. In rural area 402 with population density of 0-500 person per square kilometer, the eye's size for searching a coffee shop may be a few kilometers or more. The higher local population density, the smaller the size of the eye for local search is.
Fig. 5 is a block diagram illustrating an exemplary environment in which embodiments of the present invention may operate. In one implementation, the client side is a desktop computer or laptop which the customer uses to browse web pages. In another implementation, the client side is a mobile device such as smart phone or tablet PC
which the customer uses to find the nearby services. The preferred user interface on the client side is a web browser. The client side and the search site can be operated in an internet or intranet environment. The client 501 send the query 502 with location information to the search site 503 which contains the location based search engine 504.
The location based search engine 504 divide the search area into the eye and the band, calculates weights for topic relevance, item's importance and location within the two areas and ranks the websites or business list which corresponds to user's query 502 and send the results back to the client 501.
Fig. 6 is a flow diagram of an exemplary process for using Typhoon Model to rank websites or business list. Firstly, identify location for searching 610. In one
Fig. 3 is the exemplary diagram showing how the size of the eye is impacted by user's transportation tools. When user walks 301, the size of the eye can be just a few hundred meters in radius for searching "restaurant". When user rides a bicycle 302, the size of the eye can be one or two kilometers in radius for searching "restaurant".
When user drives a car 303, the size of the eye can be ten kilometers or more in radius for searching "restaurant". The faster the transportation tool, the larger the size of the eye is.
Fig. 4 is the exemplary diagram showing how the size of the eye is impacted by local population density. In downtown area 401 of Toronto with population density of 10000 people per square kilometer, the eye's size for searching a coffee shop may be a few hundreds of meters. In rural area 402 with population density of 0-500 person per square kilometer, the eye's size for searching a coffee shop may be a few kilometers or more. The higher local population density, the smaller the size of the eye for local search is.
Fig. 5 is a block diagram illustrating an exemplary environment in which embodiments of the present invention may operate. In one implementation, the client side is a desktop computer or laptop which the customer uses to browse web pages. In another implementation, the client side is a mobile device such as smart phone or tablet PC
which the customer uses to find the nearby services. The preferred user interface on the client side is a web browser. The client side and the search site can be operated in an internet or intranet environment. The client 501 send the query 502 with location information to the search site 503 which contains the location based search engine 504.
The location based search engine 504 divide the search area into the eye and the band, calculates weights for topic relevance, item's importance and location within the two areas and ranks the websites or business list which corresponds to user's query 502 and send the results back to the client 501.
Fig. 6 is a flow diagram of an exemplary process for using Typhoon Model to rank websites or business list. Firstly, identify location for searching 610. In one
5 implementation, the location information is longitude and latitude from GPS in client's equipment such as smart phone and tablet. In another implementation, the location information is a city from Geolocation based on client's IP address or geographical name in query text. Secondly, identify topics for searching and assess geo-sensitivity of the query 620. Topics are from user's input and these topics' geo-sensitivity can be assessed by a lookup table. Thirdly, determine the size of the eye for searching 630 and calculate the scores of the websites or business list within the eye 635. The size of the eye can be determined by one or a combination of the following factors: query's geo-sensitivity, user's transportation tools and local population density; the retrieval scores of items can be calculated by a function which consist of topic relevance weight, website's or business's importance weight and location weight. Within the eye, location weight will be assigned a constant. Fourthly, determine the size of the band for searching 640 and calculate the scores of the websites or business list within the band 645; the size of the band is proportional to the size of the eye, and for example 5-10 times of the eye in radius. Within the band, location weight will be a decreasing function of distance between query's location and business or website's location. Finally, Rank the websites or business list within the whole search area 650. The ranking will be based on the retrieval scores of websites or business list.
Fig. 7 is the exemplary diagram showing comparison of local search results ranked by Typhoon Model and distance measurement. The results for searching local restaurants ranked by Typhoon Model are shown on the left side 710. The parameters for Typhoon model 715 are as follows: centre is located at 43.67N and 79.39W and the eye for searching is 1000 meter in radius. The results for searching local restaurants ranked by distance are shown on the right side 720. The parameters for distance model 725 are as follows: The centre is located at 43.67N and 79.39W. Compared these two lists, three (kegsteakhouse.com, fransrestaurant.com and thinkglobaleatlocal.ca) of top five websites ranked by Typhoon model is not in the top 10 list ranked by distance model because they are a little far away. But they are outstanding in the top 5 in Typhoon Model due to other factors such as more references from other websites or better reviews.
Fig. 7 is the exemplary diagram showing comparison of local search results ranked by Typhoon Model and distance measurement. The results for searching local restaurants ranked by Typhoon Model are shown on the left side 710. The parameters for Typhoon model 715 are as follows: centre is located at 43.67N and 79.39W and the eye for searching is 1000 meter in radius. The results for searching local restaurants ranked by distance are shown on the right side 720. The parameters for distance model 725 are as follows: The centre is located at 43.67N and 79.39W. Compared these two lists, three (kegsteakhouse.com, fransrestaurant.com and thinkglobaleatlocal.ca) of top five websites ranked by Typhoon model is not in the top 10 list ranked by distance model because they are a little far away. But they are outstanding in the top 5 in Typhoon Model due to other factors such as more references from other websites or better reviews.
6 The foregoing description provides illustrations of preferred embodiments of the present invention, but is not intended to be exhaustive or limiting the invention to the precise form disclosed. The modifications and variations are possible by learning from the above description or may be acquired through practice of the invention.
For example, factors impacting the size of the eye have been described with regard to Fig. 2, Fig.3 and Fig. 4. There may be more factors, such as user's condition and preference and item's spatial pattern impacting the size of the eye. These factors can be used in embodiments separately or combined with previously described factors.
The preferred mode for carrying out the invention is a client/server environment described in Fig. 5. The invention can also be implemented in a single device such as Car Navigation System if the database is not very large.
The search Items in previous descriptions are websites or business lists.
However, the invention is suitable for searching any local items stored in relational database or linked database, such as local events and local goods.
For example, factors impacting the size of the eye have been described with regard to Fig. 2, Fig.3 and Fig. 4. There may be more factors, such as user's condition and preference and item's spatial pattern impacting the size of the eye. These factors can be used in embodiments separately or combined with previously described factors.
The preferred mode for carrying out the invention is a client/server environment described in Fig. 5. The invention can also be implemented in a single device such as Car Navigation System if the database is not very large.
The search Items in previous descriptions are websites or business lists.
However, the invention is suitable for searching any local items stored in relational database or linked database, such as local events and local goods.
7
Claims (19)
1. A method of ranking local search items, comprising the steps of:
dividing the search area into two parts: the eye and the band for local search;
determining location weight within the eye which is not impacted by distance between the search centre and items;
calculating location weight within the band based on distance between the search centre and items; and, ranking the websites or business list within the search area based on the retrieval scores which combine location weight, topic relevance weight, and weight for importance of websites or business list.
dividing the search area into two parts: the eye and the band for local search;
determining location weight within the eye which is not impacted by distance between the search centre and items;
calculating location weight within the band based on distance between the search centre and items; and, ranking the websites or business list within the search area based on the retrieval scores which combine location weight, topic relevance weight, and weight for importance of websites or business list.
2. The method of claim 1, wherein the boundaries of the eye and the band are circles having same centre. The eye is the inner area enclosed by the inner circle, and the band is the area between the inner circle and the outer circle.
3. The method of claim 1, wherein the size of the eye is determined by one or a combination of the following factors: query's geo-sensitivity, user's transportation tools and local population density.
4. The method of claim 3, wherein the size of the eye is a function of topic's geo-sensitivity. The higher the topic's geo-sensitivity, the smaller the size of the eye is.
5. The method of claim 4, wherein the function for the size of the eye and topic's geo-sensitivity is represented by a lookup table.
6. The method of claim 3, wherein the size of the eye is a function of transportation tools that client can use. The faster the transportation tool, the larger the size of the eye is.
7. The method of claim 3, wherein the size of the eye is a function of local population density. The higher the population density, the smaller the size of the eye is.
8. The method of claim 1, wherein the size of the band is proportional to the size of the eye. The larger the eye, the larger the size of the band for search is.
9. The method of claim 1, wherein the size of the band is 5 to 10 times larger than the size of the eye in radius.
10. The method of claim 1, wherein the location weight within the eye is a constant.
11. The method of claim 1, wherein the location weight within the band is a function of the distance between search centre and item's location.
12. The method of claim 11, wherein the location weight decreases as the distance increases within the band.
13. The method of claim 1, wherein the location weight within the eye is greater than or equal to the highest location weight within the band.
14. The method of claim 1, wherein the eye of local search area moves automatically as device in client side moves.
15. The method of claim 1, wherein the eye of local search area moves as user moves the search centre manually.
16. A computer-readable medium that stores instructions executable by one or more processing devices to perform a method for improving local search performance, comprising:
instructions for dividing the search area into two parts: the eye and the band for local search;
instructions for determining location weight within the eye which is not impacted by distance between the search centre and items;
instructions for calculating location weight within the band based on distance between the search centre and items; and, instructions for ranking the websites or business list within the search area based on the retrieval scores which combine location weight, topic relevance weight, and weight for importance of websites or business list.
instructions for dividing the search area into two parts: the eye and the band for local search;
instructions for determining location weight within the eye which is not impacted by distance between the search centre and items;
instructions for calculating location weight within the band based on distance between the search centre and items; and, instructions for ranking the websites or business list within the search area based on the retrieval scores which combine location weight, topic relevance weight, and weight for importance of websites or business list.
17. A computer-readable medium of claim 16, further comprising:
one or more instructions for determining the size of the eye by query's geo-sensitivity.
one or more instructions for determining the size of the eye by query's geo-sensitivity.
18. A computer-readable medium of claim 16, further comprising:
one or more instructions for determining the size of the eye by user's transportation tools.
one or more instructions for determining the size of the eye by user's transportation tools.
19. A computer-readable medium of claim 16, further comprising:
one or more instructions for determining the size of the eye by local population density.
one or more instructions for determining the size of the eye by local population density.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CA2821895A CA2821895A1 (en) | 2013-07-29 | 2013-07-29 | Improving local search ranking using typhoon model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CA2821895A CA2821895A1 (en) | 2013-07-29 | 2013-07-29 | Improving local search ranking using typhoon model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CA2821895A1 true CA2821895A1 (en) | 2015-01-29 |
Family
ID=52471832
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA2821895A Abandoned CA2821895A1 (en) | 2013-07-29 | 2013-07-29 | Improving local search ranking using typhoon model |
Country Status (1)
| Country | Link |
|---|---|
| CA (1) | CA2821895A1 (en) |
-
2013
- 2013-07-29 CA CA2821895A patent/CA2821895A1/en not_active Abandoned
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11461336B2 (en) | Selecting between global and location-specific search results | |
| US20220014585A1 (en) | Providing geocoded targeted web content | |
| US8301639B1 (en) | Location based query suggestion | |
| US10915580B2 (en) | Providing query suggestions and/or search results based on determined user activity | |
| US9430510B2 (en) | Computer implemented methods and systems for multi-level geographic query | |
| US8463772B1 (en) | Varied-importance proximity values | |
| US9275154B2 (en) | Context-sensitive point of interest retrieval | |
| US9384291B2 (en) | Generating geographical keywords for geotargeting search engine-offered advertisements | |
| US9194716B1 (en) | Point of interest category ranking | |
| US8538973B1 (en) | Directions-based ranking of places returned by local search queries | |
| US9143541B1 (en) | Systems, computer-implemented methods, and computer-readable media to target internet-based services on a geographic location | |
| US8898173B1 (en) | Ranking location search results based on multiple distance measures | |
| US8332396B1 (en) | Resource geotopicality measures | |
| US10146883B2 (en) | Determining labels from similar geographic features | |
| US10102222B2 (en) | Semantic geotokens | |
| US8495046B1 (en) | Encoding locations and using distances for resources | |
| US8738602B1 (en) | Determining relevance scores for locations | |
| US9715553B1 (en) | Point of interest retrieval | |
| Zou et al. | A context-aware recommendation system using smartphone sensors | |
| US20150227583A1 (en) | Managing search results | |
| NL2008665C2 (en) | Retrieving ratable content based on a geographic location. | |
| US20130268540A1 (en) | Biasing geocoding of queries | |
| US8538944B1 (en) | Resource catchment areas | |
| CA2821895A1 (en) | Improving local search ranking using typhoon model | |
| US9116996B1 (en) | Reverse question answering |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| FZDE | Dead |
Effective date: 20190730 |