Patent Application
20120271541
The invention relates to the field of route recommendations.
For any given route from a first location to a second location, there usually exists a number of acceptable alternative routes. For example, consider a person (e.g., Johan) who lives in an apartment building and walks to a nearby train station each morning as part of his daily commute to work. Johan may have several acceptable routes to choose from when walking to the train station. If we assume Johan has only recently moved to the apartment building, Johan may not realize that alternative routes to the train station exist. Moreover, even if he is aware of these alternative routes, he may not realize that some of these routes have certain advantages over the route he usually takes each morning.
What is desired, therefore, are systems and methods for providing route recommendations.
A novel route recommendation apparatus and method is described herein. The route recommendation apparatus, at least in some embodiments, functions to determine, for a user, a set of one or more recommended routes going from a starting location to an ending location in response to receiving from the user a route request message identifying the starting and ending locations. Advantageously, in some embodiments, the route recommender apparatus determines a recommended route by analyzing a set of routes between the starting and ending locations, where each route in the set was traversed by a person who is in a “social graph” belonging to the user (e.g., each route in the set was traversed by one of the user's Facebook friends). In this way, routes can be recommended to the user based on routes that are popular with the user's friends. If a user were to follow such a recommended route, the probability that the user will run into a friend or an object a friend found interesting will likely be higher than if the user selected a route using some other criteria.
Accordingly, in one aspect, a route recommender apparatus for recommending a route to a user is provided. In some embodiments, the route recommender apparatus includes a data storage system for storing, for each of a plurality of persons, location information associated with the person, where the location information may include information identifying places that the person has previously visited or passed through. The route recommender apparatus also includes a network interface for receiving route request messages. The route recommender apparatus further includes a data processing system configured to perform a route recommendation method in response to the network interface receiving a route request message that was transmitted by a communication device operated by a user, where the route request message requests a recommended route from a first location to a second location.
In some embodiments, this route recommendation method includes the steps of: (a) determining a recommended route going from the first location to the second location, wherein the determining step comprises using at least some of the stored location information in determining the recommended route; and (b) using the network interface to transmit to the communication device a route request response message comprising information identifying the determined recommended route. The information identifying the determined recommended route may include information identifying recommended places to visit—i.e., points-of-interest. The information identifying the recommended places to visit may include information identifying an implicit or explicit rating one of the plurality of persons gave to one of the recommended places to visit.
In some embodiments, the method further includes determining a set of persons who are in a social graph belonging to the user, and the step of using at least some of the stored location information in determining the recommended route comprises using only stored location information that is associated with a person who is in the user's social graph in determining the recommended route.
In some embodiments, the location information associated with one of the plurality of persons includes information identifying places the person has previously visited and may further include the time at which the person visited those places. The location information associated with the one of the plurality of persons may further include rating information identifying a rating the person gave to one of the places the person visited. In some embodiments, the step of using at least some of the stored location information in determining the recommended route includes using the information identifying a route going from the first location to the second location traversed by said person and the rating information.
In some embodiments, the location information associated with one of the plurality of persons further includes time information identifying a time of day at which the person traversed the route. In such embodiments, the method may further include determining the current time of day, and the step of using at least some of the stored location information in determining the recommended route comprises using the time information and the determined current time of day in determining the recommended route.
In some embodiments, the data storage system is operable to store preference information belonging to the user, and the step of determining the recommended route further comprises using the preference information in determining the recommended route.
In another aspect, a route recommendation method for recommending a route to a user is provided. In some embodiments, the route recommendation method includes the step of, for each of a plurality of persons, storing location information associated with the person, where the location information includes information identifying places that the person has previously visited or passed through. The method also includes the step of receiving, at a server, a route request message transmitted from a communication device operated by the user, where the route request message requests a recommended route from a first location to a second location. The method further includes the step of determining a recommended route going from the first location to the second location in response to receiving to receiving the route request message. This determining step includes using at least some of the stored location information in determining the recommended route. The method also includes the step of transmitting, from the server and to the communication device, a route request response message comprising information identifying the determined recommended route.
The above and other aspects and embodiments are described below with reference to the accompanying drawings.
The accompanying drawings, which are incorporated herein and form part of the specification, illustrate various embodiments of the present invention and, together with the description, further serve to explain the principles of the invention and to enable a person skilled in the pertinent art to make and use the invention. In the drawings, like reference numbers indicate identical or functionally similar elements.
Referring now to
In determining the recommended route for user 101 in response to receiving route request message 191, route recommender apparatus 106 may use, for each of a plurality of persons, stored location information associated with the person. The stored location information may identify one or more routes the person has traversed within the last X days, weeks, months, where X>0. Advantageously, in some embodiments, route recommender apparatus 106 determines a recommended route by: (a) determining a set of persons who are in a “social graph belonging to user 101” and (b) for each person included in the determined set, using stored location information that is associated with the person in determining the recommended route.
As used herein, a “social graph belonging to user 101” is a set of persons who are directly or indirectly associated with user 101. As an example, persons directly associated with user 101 may include all persons that user 101 has identified as being a friend of user 101, and a person indirectly associated with user 101 may include friends of user 101's friends. As a particular example, if user 101 has a Facebook account, then the social graph belonging to user 101 may include all of persons who are Facebook “friends” with user 101, and, in some embodiments, user 101's social graph may further include all persons who are a Facebook friend with a Facebook friend of user 101. In yet other embodiments, if user 101 has a Twitter account, then user 101's social graph may also include all persons who are user 101's Twitter followers (i.e., are directly associated with user 101 on Twitter). In some embodiments, user 101's social graph may also include persons who share one more traits with user 101 (e.g., age, location, sex, marital status, income, hobbies, preferences). These persons are indirectly associated with user 101.
As a specific example, assume user 101 has sent to route recommender apparatus 106 a route request message 191 requesting a recommended route from user 101's apartment building to the nearest metro (or subway) stop. In response to receiving such a route request message 191, route recommender apparatus 106 may (a) determine persons who are Facebook friends with user 101, (b) for each such person, (b1) retrieve from a location information database 202 (see
Referring now to
Route selector 208 may use the information identifying user 101 to obtain one or more other user identifiers associated with user 101. For example, route selector 208 may use the information identifying user 101 to obtain user 101's user identifier and password for each of Facebook, Twitter, LinkedIn, etc. This information may be maintained in username database 204, which may include one or more tables mapping information identifying user 101 with a set of one or more user identifiers and passwords belonging to user 101 for external social networking systems.
The obtained user identifiers may then be used by a social graph module 206 to determine a social graph belonging to user 101. Social graph module 206 may determine at least a part of user 101's social graph by transmitting to an external social network server 112 (e.g., a Facebook server or a LinkedIn server) a request message for the purpose of obtaining from server 112 the identities (e.g., names or user identifiers) of people that are directly and indirectly associated with user 101. For instance, if server 112 is a Facebook server, then the request message may include user 101's Facebook user identifier and may request server 112 to transmit to social graph module 206 a response message containing a list of user 101's Facebook friends and a list of the people that are friends with user 101's Facebook friends.
In response to social graph module 206 determining user 101's social graph, or portion thereof, (i.e., determining persons that are directly or indirectly associated with user 101), route selector 208, for each person included in at least a subset of the persons that were determined to be directly or indirectly associated with user 101, obtains from location information database 202 location information associated with the person so that route selector 208 can determine routes the person has traversed.
Referring now to
In any event, location module 210 may be configured to periodically (e.g., every 5 minutes, 30 minutes, 2 hours, etc) obtain person X's (as well as a plurality of other persons') current location. If person X's current location has changed since the last time location module 210 determined person X's current location, then location module 210 may add a row to the table associated with person X's current location (and if such a table does not exist, location module 210 may create the table and add a row to the table). As shown in
Because location information database 202 may contain, for each of a plurality of persons, a table, like table 300, for each location the person visits or passes through, information database 202 contains information identifying routes that each of the plurality of persons has traversed. Thus, when user 101 requests a recommended route, route selector 208 can use information obtained from social graph module 206 (e.g., the list of user 101's “friends”) and from location information database 202 to determine routes traversed by user 101's friends. With this route information in hand, route selector 208 can perform any one of a variety of algorithms to determine a recommended route using, among other things, the obtained route information for the people within user 101's social graph. For example, in some embodiments, route selector 208 may use a well known “ant colony optimization (ACO)” algorithm.
After determining a recommended route, route selector 208 may provide information identifying the determined recommended route to point-of-interest module 214, which may be configured to determine recommended points-of-interest for user 101 that can be found on (or near) the determined recommended route. For example, using location information from location information database 202, point-of-interest module 214 may be able to determine that many of the people within user 101's social graph who traverse the recommended route regularly stop at a particular location on the route (e.g., a coffee shop). Additionally, point-of-interest module 214 may be able to determine from information stored in user preference information database 216 that user 101 is a coffee lover. Using this information, point-of-interest module 214 can be configured to conclude that the coffee shop is a recommended point-of-interest for user 101 and add to the information identifying the determined recommended route point-of-interest information for one or more of the recommended points-of-interest along the route.
In some embodiments, in response to determining that a particular location is a point-of-interest for user 101, point-of-interest module 214 may transmit to a point-of-interest server 114 a request message identifying the determined particular location and requesting information concerning the location. For example, if the location is a coffee shop, then point-of-interest server 114 may respond to the request message by transmitting to point-of-interest module 212 a response message containing information about the coffee shop (e.g., street address, hours of operation, phone number, web address, review information, or advertising). In response to receiving the information about the coffee shop, point-of-interest module 212 may add the information about the coffee shop to the information identifying the determined recommended route that was received from route selector 208. Point-of-interest module 212 may then pass the updated information identifying the determined recommended route to an output generator 214.
In response to receiving from point-of-interest module 212 the updated information identifying the determined recommended route, output generator 214 may request from an advertising server 120 an advertisement to include with the information received from point-of-interest module 212. Additionally, output generator 214 may be configured to obtain user 101's display preferences from user preference information database 216 and format the information received from point-of-interest module 212 is a display format preferred by user 101. Once the information received from point-of-interest module 212 (i.e., the information identifying the determined recommended route and any point-of-interest information) has been formatted, output generator 214 transmits to user 101's communication device 102 the route request response message 192 discussed above. After receiving route request response message 192, a route recommender application 288 executing on user 101's communication device 102 may display on a display 289 of communication device 102 a map with the determined recommended route and any of the points-of-interest determined by point-of-interest module 212 highlighted so that user 101 can easily see the recommended route and points-of-interest along that route.
In some embodiments, in response to a route request message, route selector 208 may determine a set of recommended routes instead of a single recommended route and the route request response message 192 may include, for each recommended route, route information identifying the determined recommended route.
Referring now to
In some embodiments, after route recommender apparatus 106, in response to a route request message 191 received from user 101's communication device 102, determines a set of one or more recommended routes and transmits a route request response message 192 to communication device 102, route recommender apparatus 106 may periodically (and automatically) update the set of recommended routes and transmit to communication device 102 updated route information. For example, after route recommender apparatus 106 determines an initial set of one or more recommended routes in response to receiving a route request message 191, route recommender apparatus 106 may periodically determine whether routes should be added and/or removed from the initial set of recommended routes and also whether points-of-interest should be added and/or removed from a route included in the initial set of recommended routes.
For example, if route recommender apparatus 106, based on location information from location information database 202, determines that one of the routes that was initially recommended is no longer being used by user 101's friends (or is being used only infrequently), the route recommender apparatus 106, in response to this determination, may transmit to communication device 102 a message indicating that the route is no longer a recommended route. As another example, if route recommender apparatus 106, based on location information from location information database 202, determines that at least a certain percentage of the persons in user 101's social graph who traverse a route included in the initial set of routes stop at a new location located on the route, then route recommender apparatus 106, in response to this determination, may transmit to communication device 102 a message indicating that the route has a new point-of-interest and including information pertaining to this new point of interest (e.g., its location, among other things) so that when user 101 uses route recommender application 288 to view the route, route recommender application 288 can display on the route a star shaped object correspond to the newly discovered point of interest.
Referring now to
In step 504, route recommender apparatus 106 receives from user 101's communication device 102 a route request message 191. The route request message identifies a starting location and an ending location and requests a recommended route from the starting location to the ending location.
In step 506, route recommender apparatus 106 determines a user identifier for user 101. For example, route recommender apparatus 106 may determine the MN of communication device 102 or the route request message 191 itself may contain the user identifier for user 101.
In step 510, route recommender apparatus 106 uses the user identifier determined in step 506 to determine a set of persons who are in user 101's social graph. As described above, route recommender apparatus 106 may include a username database that maps the user identifier determined in step 506 with a set of one or more other user identifiers (e.g., user 101's Facebook user identifier) that can be used to obtain a list of persons who are directly or indirectly associated with user 101.
In step 512, for each person included in the set of persons determined in step 510, route recommender apparatus 106 retrieves location information associated with the person and, using the retrieved location information, determines whether the person traversed a route going from the starting location to the ending location. That is, in step 512, route recommender apparatus 106 determines a set of routes going from the starting location to the ending location, where each route in the set was traversed by at least one of the persons included in the set of persons determined in step 510.
In some embodiments, in step 512, route recommender apparatus 106 takes into account the current time of day and the times that the determined routes were traversed. For example, in some embodiments, route recommender apparatus 106 includes in the determined set of routes only those routes that were traversed at about the same time of day as when the route request message was received. So, for example, if the route request message is received in the morning, route recommender apparatus 106 will include in the set of determined routes only those routes that were traversed in the morning time. As illustrated in
In step 514, route recommender apparatus 106 analyzes the set of routes from the starting location to the ending location determined in step 512 to determine a set of one or more recommended routes.
In step 516, route recommender apparatus 106 designates certain places on a recommended route as points-of-interest
In step 518, route recommender apparatus 106 provides to user 101 (e.g., transmits to user 101's communication device 102) information identifying the determined recommended route(s) and the designated points-of-interest.
Referring now to
In step 602, route recommender apparatus 106 determines a recommended route going from the starting location to the ending location. For example, in step 602, route recommender apparatus 106 may use conventional algorithms to determine the fastest or shortest route from the starting location to the ending location.
In step 604, for each person included in the set of persons determined in step 510, route recommender apparatus 106 retrieves location information associated with the person and, using the retrieved location information, determines places the person has visited on or near the determined recommended route and the person's affinity (i.e., liking) for the place. There are several way that route recommender apparatus 106 can determine the person's affinity for a place. First, if the person has rated the place (e.g., given the place four out four stars), then route recommender apparatus 106 can determine the person's affinity simply by determining the rating the person gave to the place. Another way route recommender apparatus 106 can determine the person' affinity for a place is to determine how often the person has visited the place within some arbitrary period of time (e.g., the last six months). If the person visits the place every day, that is an indication that the person has a high liking for (i.e., affinity for) the place, whereas if the person visited the place only once and never returned, then that is an indication that the person has no or low affinity for the place.
In step 606, route recommender apparatus 106 designates certain places on the recommended route as points-of-interest based on the place affinity information determined in step 604. For example, if route recommender apparatus 106 determines that several of the people within user 101's social graph have a high affinity for a place that is on or near the recommended route, then route recommender apparatus 106 may designate that place as a point-of-interest.
In step 608, route recommender apparatus 106 provides to user 101 (e.g., transmits to user 101's communication device 102) information identifying the determined recommended route(s) and the designated points-of-interest.
Referring now to
Referring now to
Johan is a trucker who recently started to drive from Washington to New Orleans. Johan is Facebook friends with five other truckers, who also travel from Washington to New Orleans. Let us assume there are 50 different possible routes from Washington to New Orleans, all of equal convenience. Hence, each of the five friends may take different routes every time they go from Washington to New Orleans (and back). Johan wants to find the most interesting route to go from Washington to New Orleans, and where the best coffee is along the route. Johan, therefore, sends a route request message 191 to route recommender apparatus 106. In response, route recommender apparatus 106 determines Johan's Facebook friends and the routes these friends have taken from Washington to New Orleans over the past few months (an arbitrary, but user-definable, period). Route recommender apparatus 106 then collates the instances of the different routes that the five trucker friends have taken over the period of time, and the instances when they have stopped at a location. The frequency of the routes taken and the locations stopped at are used to determine a recommended route and points-of-interest along the route. Locations where a friend made a note may be given an additional weight. After determining a recommended route and points-of-interest, route recommender apparatus 106 may then transmit to point-of-interest server 114 a request message identifying the determined point-of-interests and requesting information concerning these locations. After route recommender apparatus 106 receives a response from server 114, route recommender apparatus 106 then generates a route request response message 192 that will include information identifying the determined recommended route(s) and information pertaining to the determined points-of-interest, and transmits the response message to Johan. The response message may further include information identifying the times of day when Johan's friends typically depart Washington for New Orleans. In response to receiving the route request response message, Johan's communication device may display the recommended route(s) and recommended points-of-interest on a map as shown in
When Johan wants to go to New Orleans, Johan may leave at the time of day indicated in the response message and follow the recommended route. Johan may stop for a coffee at places that one of his friends has rated highly. Hence, the probability that Johan will meet a friend while traversing the recommended route is greater than if Johan selected one of the 50 routes at random, and the probability that Johan will get better coffee than otherwise is also larger.
While various embodiments of the present invention have been described above, it should be understood that they have been presented by way of example only, and not limitation. Thus, the breadth and scope of the present invention should not be limited by any of the above-described exemplary embodiments. Moreover, any combination of the above-described elements in all possible variations thereof is encompassed by the invention unless otherwise indicated herein or otherwise clearly contradicted by context.
Additionally, while the processes described above and illustrated in the drawings are shown as a sequence of steps, this was done solely for the sake of illustration. Accordingly, it is contemplated that some steps may be added, some steps may be omitted, the order of the steps may be re-arranged, and some steps may be performed in parallel.
