KR20150015259A - Method for creation of recommend POI and Sever - Google Patents

Abstract

The purpose of the present invention is to provide a method for generating a recommended POI to be recommended to a user when a search POI searched by the user is not searched. The present invention relates to a method for generating a recommend POI and a POI recommending server supporting the same and, more specifically, to a method for generating a recommend POI by using a weighting on collected information and a POI recommending server supporting the same. To achieve this, the POI recommending server includes a reference storage unit configured to store an initially searched search POI, a search POI searched after the initial search when a search of the initially searched search POI has failed, and a type of business of the search POI; a search POI utilization storage unit configured to store whether to guide a user to a search POI of which the search is successful and whether a route to the search POI has been provided to the user; and a recommended POI generation unit configured to generate a recommended POI for a search POI of which the search has failed by using information stored in the reference storage unit and information stored in the search POI utilization storage unit.

Description

[0001] The present invention relates to a method for generating a recommended POI,

The present invention relates to a method for generating a recommended POI and a POI recommendation server supporting the same, and more particularly, to a POI recommendation server for generating or supporting a recommended POI using a weight for collected information.

In the conventional navigation search matching system, an information search using a name provides an index file defined in advance in an administrative code and a dictionary sequence by using a search sort filter specified in advance in a basic search database. When a user inputs a Hangul character collection for searching for a name, the user searches for a pre-stored index file in a top-down manner and provides it.

Korean Patent Laid-Open Publication No. 2009-0117355 (a title of the invention: method and apparatus for updating a map of a navigation system) proposes a map updating method of a navigation terminal. In particular, when searching for a POI And registering the logged POI that failed to search as a nickname candidate of the reentered POI when the destination specified by the navigation and the destination where the user arrived are the same by the re-input POI.

Korean Patent Publication No. 2007-0106217 (Navigation Matching System of Navigation) is a database for storing a history of user's path selection and an index file generating unit for accumulating search terms accumulated by a user and making database of search terms having a large frequency, A manager module module for defining a rule to be applied to the priority among these, and a user for searching for information such as a name search using a navigation system, when there is a matching candidate group according to an available keyword, As a module in charge, the location to be displayed is matched to a matching candidate part displayed in association with the Korean character input window for searching for a name, and automatically selected for a candidate string selected after the user selects a candidate group. And a display unit which is an LCD screen of navigation It proposes the Matching system of navigation that castle.

Generally, when there is no specific POI requested by the user, it is common to recommend the recommended POI corresponding to the specific POI to the user by using the frequency. However, when recommending a recommendation POI to a specific POI only by using only the frequency, it is general that the relevance to the POI to be searched by the actual user is low. Therefore, it is necessary to recommend a recommendation POI to the user which substantially conforms to the user's request.

A problem to be solved by the present invention relates to a method for generating a recommendation POI to be recommended to a user when a search POI searched by a user is not searched.

Another problem to be solved by the present invention relates to a method of providing accurate information on a commercial entity to a user who wants to analyze commercial property of the building.

Another problem to be solved by the present invention relates to a method for generating a recommendation POI by using various information and a weight for the information.

To this end, the POI recommendation server of the present invention includes a search POI that is searched for the first time, a search POI that is searched in the case that the search POI that was searched for the first time is failed, a preference storage unit that stores a business type of the search POI, A retrieval POI utilization storage unit which stores whether or not a path has been provided up to the retrieval POI provided with the path; a recommendation POI for the retrieval POI that has failed to retrieve information stored in the preference storage unit and information stored in the retrieval POI utilization storage unit; And a recommended POI generation unit for generating a recommended POI.

To this end, the method for generating a recommended POI of the present invention includes storing a search POI that is searched first, a search POI that is searched for in the first search POI, a search POI type in the preference storage unit, A search POI utilization storage unit for storing a search POI using the information stored in the preference storage unit and information stored in the search POI utilization storage unit; Lt; RTI ID = 0.0 > POI < / RTI >

The recommendation POI generation method and the POI recommendation server supporting the POI recommendation server according to the present invention are designed such that the first POI inputted by the user is the POI which is the most preferred by the user And it is effective to increase the sales by inducing the store of the POI having the POI by utilizing it in the analysis of the commercial area in the future.

Also, it is possible to improve the satisfaction of the user by providing a recommendation POI for the failed POI based on various conditions such as the weight of the repeatedly inputted search POI and whether the actual guidance is provided until the search POI.

FIG. 1 illustrates a process of generating a recommendation POI based on a search word retrieved by a user according to an embodiment of the present invention.
FIG. 2 is a diagram illustrating an operation performed when a search for a search POI fails. FIG.
FIG. 3 illustrates an example in which different weights are assigned according to a search order of search POIs requested by a user according to an embodiment of the present invention.
FIG. 4 illustrates a configuration for generating a recommended POI from a terminal (or server) according to an embodiment of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS The foregoing and further aspects of the present invention will become more apparent from the following detailed description of preferred embodiments with reference to the accompanying drawings. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Reference will now be made in detail to the embodiments of the present invention, examples of which are illustrated in the accompanying drawings.

FIG. 1 illustrates a process of generating a recommendation POI based on a search word retrieved by a user according to an embodiment of the present invention. Hereinafter, a process of generating a recommendation POI based on a search word retrieved by a user according to an embodiment of the present invention will be described in detail with reference to FIG.

The generation of the recommended POI can be performed by either the user's terminal or the POI recommendation server receiving the information from the terminal.

In step S100, the user inputs a search POI (search term) to be searched in the terminal.

In step S110, if there is no search POI requested by the user, the terminal moves to step S120. If there is a search POI requested by the user, the terminal moves to step S130. The operation performed in step S120 will be described with reference to FIG. That is, the UE performs a different operation according to the number of searches of the POI requested by the user, and generates a recommended POI according to the result of the search.

In step S130, the terminal displays the path to the search POI requested by the user using a display unit associated with the terminal or the terminal.

In step S140, the terminal determines whether the user has reached the search POI displayed by the user. That is, the user can move to the corresponding point using the search POI or stop the route guidance request in the middle without moving to the search POI. As described above, the present invention determines whether the search POI requested by the user has actually arrived.

In step S150, the terminal checks the distribution of APs used to provide the path to the search POI (destination).

In step S160, the terminal reconstructs a recommendation POI based on the collected information. That is, the terminal reconfigures the recommendation POI using the search distribution of the search POI requested by the user, reaching the search POI provided, and the AP distribution located on the route or the search POI. Of course, the reconfiguration of the recommendation POI can be performed on an external POI recommendation server other than the terminal.

Hereinafter, a case where the search fails for the search POI will be described with reference to FIG.

If there is no search result for the search POI requested by the user in step S200, the search POI is stored in the preference POI storage unit. The terminal determines that the POI that the user is aware of in the case of the search POI requested by the first user. However, if there is no POI requested by the user, the search POI is determined to be the preferred POI by the user and stored in the preferred POI storage unit. If the search POI is not found, the search POI and log information are stored in the server.

For example, in order to visit a coffee shop located in a specific building, a coffee shop that a user frequently visits is set as a search POI, and a search request is made. Hereinafter, the search POI requested by the user for the first search is called "AAAA ". Of course, in this case, the user can recognize that there is a corresponding coffee shop in the building, but he can ask for a search without recognizing it.

Also, the search POI does not exist in the interior of the building but can exist outside the building. That is, when an outdoor search engine is utilized instead of an indoor search engine, if the search POI is searched, the business type of the search POI can be confirmed. In addition, if the search POI is not searched through the outdoor search engine, the morpheme for the search POI that failed the search can be analyzed to derive the business type. That is, when the search POI is inputted as 'Cafe Bennet', it can be confirmed that the industry of the search POI is 'Cafe' by analyzing the morpheme.

In step S210, the terminal receives a second search POI from the user. For example, the search POI requested by the user for the second time is referred to as "BBBB ". In this case, if there is a search POI requested for a second search, the terminal provides information about the search POI. If there is no search POI requested for the second search, the terminal indicates that the search POI is not present. Also, the terminal determines whether the search POI requested for the first search and the search POI requested for the second search are the same business type. That is, if the search POI requested for the second search is the same as the search POI requested for the first search, it is stored in correspondence with the search POI requested for the first search.

In step S220, the terminal receives a third search POI from the user. For example, the search POI requested by the user for the third time is called "CCCC ". In this case, if there is a search POI requested for a third search, the terminal provides information about the search POI. In addition, the terminal determines whether the search POI requested for the third search and the search POI requested for the second (or first) search are the same business type. That is, the search POI requested for the third time is also stored in correspondence with the search POI requested for the second (or first) search in the case of a coffee shop, the same as the search POI requested for the second (or first) search.

The above-described process is repeated to retrieve the search POI requested by the user. That is, when a user wants to visit a coffee shop located in the building, the user sequentially requests the search in order of the coffee shop preferred by the user. As described above, the present invention assigns different weights according to the search order of search POIs requested by the user. In this case, the search POI requested by the user is limited to the same company.

FIG. 3 illustrates an example in which different weights are assigned according to a search order of search POIs requested by a user according to an embodiment of the present invention.

Referring to FIG. 3, it can be seen that different weights are assigned according to the search order of the search POIs requested by the user. That is, a high weight is assigned to a search POI that has been successfully searched, and a low weight is assigned to a search POI requested for the first search. Of course, as shown in FIG. 3, a search POI that has failed to search due to an error is assigned the same weight as the search POI.

As described above, the present invention assigns different weights according to the search order of search POIs requested by the user.

FIG. 4 illustrates a configuration for generating a recommendation POI from a terminal (or a POI recommendation server) according to an embodiment of the present invention. Hereinafter, a configuration for generating a recommendation POI from a terminal (or a POI recommendation server) according to an embodiment of the present invention will be described in detail with reference to FIG.

Referring to FIG. 4, a preference storage unit 400, a search POI utilization storage unit 410, and a recommendation POI generation unit 420 are included. Of course, other configurations than the above-described configuration may be included in the configuration for generating the recommendation POI suggested by the present invention.

The preference storage unit 400 includes a preference POI storage unit 402 and a preference relevancy storage unit 404. The preference POI storage unit 402 stores information on the POIs that the user has searched relatively. The preference POI storage unit 402 stores the POI searched by the user according to various established criteria such as a business type, a time period, and a building where the user is located.

The preference relevancy storage unit 404 stores the association between the first search POI that the user first searches and the search POI that is searched for later. For example, if the first search POI and the later search POI are of the same business category as described above, the information is stored. Of course, the preference relevancy storage unit 404 stores different weights for the first retrieved POI and the later retrieved POI. That is, the weight of the retrieval POI that succeeds the actual retrieval is set lower than the weight of the retrieval POI that failed the retrieval.

The destination POI utilization storage unit 410 stores information about whether or not the route guidance for the target POI that has been searched successfully has been requested, whether the actual route guidance has been received up to the destination POI requesting the route guidance, and the AP distribution around the POI.

The recommendation POI generation unit 420 generates a recommendation POI to be recommended when the user does not search for the target POI requested by the user using the information stored in the preference storage unit 400 and the information stored in the destination POI utilization storage unit 410 . That is, the recommended POI generating unit 420 generates a recommended POI using the weight of the information stored in each storage unit. Particularly, when requesting route guidance for a target POI that is successfully retrieved from the information stored in the destination POI utilization storage unit 410, the weight is set to be higher than when the route guidance is not requested. If the route guidance is received up to the destination POI Is set to be higher than the weighted value when the vehicle is stopped from the stop (when the route guidance is not received until the destination POI). In addition, when the dense frequency of the AP is high, the weight is set to be higher than the case where the AP is dense. In this manner, the recommendation POI generator 420 of the present invention generates the recommendation POI using the information stored in the preference storage 400 and the destination POI utilization storage 410 and the weight of the information.

Of course, when the recommended POI is generated by the POI recommendation server instead of generating the recommended POI in the terminal as described above, the accuracy of the recommended POI can be increased. That is, the POI recommendation server can generate the recommended POI using the information provided from the plurality of terminals, thereby improving the accuracy compared to generating the recommended POI from the terminal.

In addition, the present invention can utilize the collected information in various ways. That is, when the user searches the terminal repeatedly for different search POIs for the same business category, the first search POI is the user's preferred POI. Therefore, if the user does not have a preferred POI in the building, he / she can recommend the shop of the merchant associated with the POI. That is, the POI recommendation server can generate and provide a preferred POI according to various criteria such as industry, age, sex, and individual.

In this case, the POI recommendation server may differ from the recommendation criteria of the recommendation POI provided to the terminal and the recommendation criteria of the recommendation POI to be provided to the terminal requesting the information about the business opportunity. That is, the recommendation POI provided to the terminal user recommends the POI actually located in the building, while the terminal requesting the information about the commercial is provided with the information on the most preferred POI even if it is not located in the actual building .

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it will be understood by those of ordinary skill in the art that various changes in form and details may be made therein without departing from the scope of the present invention .

400: preference storage unit 402: preference POI storage unit
404: Preferred affinity storage unit 410: Search POI utilization storage unit
420: Recommendation POI generating unit

Claims (12)

A preference storage unit for storing a first search POI, a first POI searched for in the first search, and a search POI and a search POI searched for in the first search;
And a recommendation POI generation unit for generating a recommendation POI for a search POI that has failed to search using information stored in the preference storage unit.
The method according to claim 1,
And a retrieval POI utilization storage unit which stores whether or not a retrieval POI that has been successfully retrieved is informed and whether or not a path to the retrieval POI has been provided.
Wherein the recommendation POI generation unit generates a recommendation POI using information stored in the search POI utilization storage unit.
3. The method of claim 2,
Wherein the POI recommendation server assigns a weight for generating the recommended POI to the first retrieved POI and the later retrieved POI if the first retrieved POI and the later retrieved POI are the same business type.
4. The POI recommendation server according to claim 3, wherein a weight that is relatively high compared to the first search POI is provided to the later-retrieved POI.
5. The apparatus according to claim 4,
And stores the AP distribution information by POI.
6. The method according to claim 5, wherein a relatively high weight is assigned to the retrieved POI that has been successfully retrieved compared to the case where the retrieval is not guided, and when the path is provided up to the retrieved POI provided with the path, And a weighting value is assigned to a POI having a relatively large distribution of APs.
7. The apparatus of claim 6,
Wherein the POI recommendation server generates a recommendation POI for the search POI that failed the search using the information stored in the preference storage unit, the information stored in the search POI utilization storage unit, and the weight of each information.
The apparatus of claim 1, wherein the recommendation POI generator,
Wherein the recommendation POI is provided differently from a POI provided for analysis of a commercial area and a recommended POI provided for a search POI that fails to be retrieved, among the POIs stored in the preference storage unit.
Storing in the preference degree storage section, a first search POI, a first search POI, and a search POI and a search POI, which are searched for in the case where the first search POI is searched;
And generating a recommendation POI for a search POI that has failed to search using information stored in the preference storage unit.
10. The method of claim 9, wherein prior to the step of generating the recommendation POI,
And storing in the retrieval POI utilization storage unit whether or not the retrieval POI that has been successfully retrieved is provided with a route to the retrieval POI provided with the route,
The step of generating the recommendation POI comprises:
Wherein the recommendation POI is generated using the information stored in the search POI utilization storage unit.
11. The method according to claim 10, wherein the step of storing in the search POI utilization storage unit comprises:
And storing the AP distribution information by the POI.
10. The method of claim 9, wherein the step of storing in the preference storage unit comprises:
Wherein a weight for generating the recommendation POI is assigned to a first search POI and a later search POI if the first search POI and the later search POI are the same business type.

Images