JP6363474B2 - Specific location discovery device and specific location discovery program - Google Patents

Description

  The present invention relates to a specific location discovery apparatus and a specific location discovery method.

  In recent years, services using location information of mobile terminals have been provided. The above position information is acquired based on, for example, positioning by a GPS function provided in the mobile terminal (for example, Patent Document 1) or the position of the radio base station to which the mobile terminal is connected.

  Web search services are also used on the Internet. In recent years, web search services via mobile terminals have been widely used along with improvements in the performance of mobile terminals and wireless communication networks.

JP 2009-140064 A

  The relationship between the place in the real space where the user has executed the search process and the place indicated by the search word (place name) searched by the user in the search process depends on the user's behavior. For example, when the user is lost in the vicinity of a certain place to be reached, it is assumed that the user performs a search process using a search term indicating the place name in the vicinity of the place.

  In addition, even when the user is located in the vicinity of a certain place, when performing a search process using the place name, it is expected that the user's behavior has some reason or tendency. It is thought that the convenience of the user will be improved by finding and specifying a place where such an action (hereinafter, referred to as “specific action”) is likely to be performed and providing a service adapted to the place. .

  In view of the above circumstances, an object of the present invention is to find a specific place where a specific action by a user is easily performed.

  The specific location finding device of the present invention includes, from a plurality of search logs, a search execution location included in the search log where a user has executed search processing, and a location name that is a search term input by the user in the search processing A candidate extraction unit that extracts candidates for a specific place that is a place where the specific action is likely to be performed, and a specific action index corresponding to the frequency at which the specific action is performed for each extracted candidate for the specific place An index calculating unit for calculating, and a specific place determining unit for determining whether the candidate is the specific place based on the calculated specific action index.

  The specific location finding method of the present invention includes, in one or more computer devices, a search execution location, which is included in the search log, from which a user has executed a search process, and a user input in the search process, from a plurality of search logs Based on a place name that is a search word, extracting a candidate for a specific place that is a place where a specific action is easily performed, and corresponding to the frequency with which the specific action is performed for each candidate of the extracted specific place Calculating a specific action index to be performed, and determining whether the candidate is the specific place based on the calculated specific action index.

  According to the present invention, a specific place which is a place where a specific action by a user is easily performed is found.

It is a figure which shows the specific location discovery server which concerns on embodiment, and the computer apparatus relevant to a specific location discovery server. It is a physical configuration block diagram of the server of the embodiment. It is a physical configuration block diagram of a user apparatus of an embodiment. It is a logical configuration block diagram of the search server of the embodiment. It is a logic block diagram of the specific location discovery server of embodiment. It is a flowchart which shows the specific place discovery operation | movement of embodiment. It is a figure which shows an example of the search log table of embodiment. It is a logical block diagram of the specific location discovery server of a modification.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

1. Embodiment 1 (1). Overview of Specific Location Discovery Server With reference to FIG. 1, a specific location discovery server DS according to an embodiment of the present invention and a computer device related to the specific location discovery server DS are schematically shown. The computer apparatus according to the present embodiment executes communication according to a predetermined protocol (TCP / IP or the like).

  The search server SS receives a search request from the user apparatus UE (for example, a smartphone) and executes a web search (acquisition of a URL list corresponding to the search request). Note that the user apparatus UE communicates with the search server SS via the radio communication network RN.

  The execution result of the web search is transmitted to the user apparatus UE as a selectable URL candidate and transmitted to the specific location discovery server DS as a search log. In the present embodiment, the search log includes a search execution place where the user has executed the search process and a search term input by the user in the search process. The search term may be a place name indicating a geographical place.

  The specific location discovery server DS performs an analysis process based on the accumulated search log, and discovers a specific location where the specific action by the user is likely to be performed (details will be described later). The specific action is a predetermined action performed by the user in the vicinity of the specific place. Non-limiting examples of specific behavior include a user getting lost at a specific location.

1 (2). Physical configuration 1 (2) -1. Configuration of Server Device FIG. 2 is a block diagram showing a physical configuration of each server device of this embodiment including the specific location discovery server DS. Each server device (DS, SS) includes a network interface 10, an input unit 12, an output unit 14, a CPU (Central Processing Unit) 16, a RAM (Random Access Memory) 18, a ROM (Read Only Memory) 20, and an HDD (Hard Disk). Drive) 22.

  The network interface 10 communicates with other computer devices via a network. The input unit 12 receives an input signal from an input device such as a keyboard. The output unit 14 transmits an output signal to an output device such as a display. The CPU 16 performs various controls and calculations by executing programs stored in the RAM 18 and the ROM 20 which are main storage devices. The HDD 22 is an auxiliary storage device that stores programs and data that can be expanded on the RAM 18. A storage medium such as an SSD (Solid State Drive) may be adopted instead of or in addition to the HDD 22.

  As a person skilled in the art understands naturally, one server may be configured by a plurality of computer devices, or one computer device may include a plurality of virtualized servers. The search server SS and the specific location discovery server DS may be configured integrally.

1 (2) -2. Configuration of User Device FIG. 3 is a block diagram illustrating a physical configuration of the user device UE of the present embodiment. The user apparatus UE includes a wireless communication interface 30, an input device 32, an output device 34, a GPS module 36, a CPU 38, a RAM 40, a ROM 42, and a flash memory 44.

  The wireless communication interface 30 communicates with other computer devices via the wireless communication network RN. The input device 32 is an element that receives an input from a user and outputs an electrical signal, and is, for example, a touch pad. The output device 34 is an element that outputs a processing result as an image, and is, for example, a liquid crystal screen. The input device 32 and the output device 34 may be configured as an integrated touch panel. The GPS module 36 acquires the latitude and longitude of the place where the user apparatus UE is located based on the radio wave from the GPS satellite. The CPU 38 performs various controls and calculations by executing programs stored in the RAM 40 and the ROM 42 which are main storage devices. The flash memory 44 is an auxiliary storage device that stores programs and data that can be expanded on the RAM 40.

1 (3). Logical configuration 1 (3) -1. Configuration of Search Server FIG. 4 is a block diagram showing a logical configuration of the search server SS. The search server SS includes a search processing unit 100, a search result presentation unit 110, and a search log transmission unit 120 as functional blocks.

  The search processing unit 100 executes search processing based on a search request from the user apparatus UE. The search request includes a search execution place (latitude and longitude) where the user has executed search processing using the user apparatus UE, and a search term input by the user.

  The search result presentation unit 110 presents the search result URL, which is the output of the search process executed by the search processing unit 100, to the user apparatus UE. The user operates the user apparatus UE to select a URL (selection URL) desired to be browsed from the search result URL. The above selection results by the user are delivered from the user apparatus UE to the search server SS.

  The search log transmission unit 120 transmits information related to the search process executed based on the search request from the user apparatus UE as a search log to the specific location discovery server DS. The search log includes a search execution location, a search word, and a selected URL selected by the user. The search log transmission unit 120 may add the selected URL selected by the user to the above search log after transmitting the search execution location and the search word as the search log when executing the search process. Further, after the search execution location, the search word, and the selected URL are prepared, they may be transmitted as a search log.

  The above functional blocks are realized by the CPU 16 executing a computer program stored in the main storage device of the search server SS.

1 (3) -2. Configuration of Specific Location Discovery Server FIG. 5 is a block diagram showing a logical configuration of the specific location discovery server DS. The specific location discovery server DS includes a candidate extraction unit 200, an index calculation unit 210, and a specific location determination unit 220 as functional blocks, and includes a search log table LT and a specific location table ST as logical tables.

  The candidate extraction unit 200 extracts candidates for a specific place that is a place where a specific action is likely to be performed, from a plurality of search logs included in the search log table LT, based on a search execution place and a search word. The index calculation unit 210 calculates a specific action index corresponding to the frequency at which the specific action is performed for each candidate at the specific place extracted by the candidate extraction unit 200. The specific place determination unit 220 determines whether or not the specific place candidate is a specific place based on the specific action index calculated by the index calculation unit 210, and stores the specific place in the specific place table ST.

  The above functional blocks are realized by the CPU 16 executing a computer program stored in the main storage device of the specific location discovery server DS. Further, the above logical table is a constituent element of a relational database in which a plurality of tables are linked based on a predetermined relation (relation). The relational database may be implemented as an on-disk database in which data is stored in the HDD 22 that is a non-volatile memory, or may be implemented as an in-memory database in which data is stored in the RAM 18 that is a volatile memory.

1 (4). Specific Location Discovery Operation The specific location discovery operation of this embodiment will be described below. FIG. 6 is a flowchart showing the specific location finding operation. In this example, “the user is lost” is adopted as an example of the specific action. Therefore, the specific place in the present example is “a place where the user frequently gets lost in the way (hereinafter, sometimes referred to as“ lost spot ”)”.

  As described above, the search log table LT stores a plurality of search logs generated as a result of the search process and including the search execution location, the search word, and the selected URL. FIG. 7 is a diagram illustrating an example of the search log table. Schematically, the specific location discovery server DS discovers a specific location (stray spot) by analyzing the search log in the search log table LT.

  The candidate extraction unit 200 executes candidate location candidate extraction processing for each search log included in the search log table LT (S1010). More specifically, when the distance between the search execution location included in one search log and the location indicated by the location name included in the search log is less than the first threshold, the candidate extraction unit 200 stores the location name. Are extracted as candidate spot spots. The candidate extraction unit 200 executes the above extraction process for a plurality of search logs in the search log table LT (that is, in units of records in the search log).

  It should be noted that the search log that is the target of the candidate location extraction process for a specific place (the lost spot) may be all the search logs in the search log table LT, or search logs within a predetermined period (for example, from the processing time point). Search logs over the past year). Further, only the search log added after the candidate extraction process executed earlier may be the target of the current candidate extraction process.

  The index calculation unit 210 calculates a specific action index (stray index) for each extracted candidate for a specific place (stray spot) (S1020). In this example, the specific action index is calculated based on the selected URL included in the search log. More specifically, the index calculation unit 210 includes, for each of the lost spot candidates extracted by the candidate extraction unit 200, a character string related to getting lost among a plurality of search logs corresponding to the lost spot candidate, That is, a search log having a selected URL including a character string (for example, access.html, map.html, etc.) indicating a page that provides a route map or guide map to a lost spot candidate is counted to calculate a lost index. The lost index may be an absolute number of counted search logs, or may be a ratio (ratio) of the counted number of search logs to the total number of search logs corresponding to the lost spot candidate. Further, the lost index may be an index based on a combination of the above absolute value and relative value. The above-described ambiguity index indicates that the greater the value, the higher the frequency of ambiguity at the ambiguity spot. In the above count, the number of search logs may be counted, or the number of search unique users may be counted. When the latter is employed, an appropriate count is performed even if a limited number of users are performing a large amount of searches.

  In other words, the index calculation unit 210 increases the frequency with which a specific action is performed on a candidate for a specific place as the number of times a selection URL including a character string related to the specific action is selected for one specific place candidate. Calculate specific behavior indicators to show that they are high. The above calculation process is executed for each candidate.

  The specific place determination unit 220 determines whether each candidate is a specific place (a lost spot) based on the calculated specific action index (a lost place index) (S1030). More specifically, the specific place determination unit 220 determines that the lost spot candidate is actually a lost spot when the lost index regarding one lost spot candidate exceeds the second threshold. The specific location determination unit 220 stores each candidate determined to be a specific location (stray spot) in the specific location table ST (S1040).

  The information on the specific place (stray spot) stored in the specific place table ST can be used arbitrarily. For example, in the route guidance service, it is assumed that the convenience of the service to the user is improved when the route guidance is performed in the vicinity of a lost spot in more detail.

1 (5). Effects of the present embodiment According to the configuration of the present embodiment described above, a specific place is found based on the search log. That is, after a specific location candidate is first extracted from a plurality of search logs, it is estimated whether or not the specific action is frequently performed in each extracted candidate (specific behavior index), and whether or not the specific location is a specific location Is determined. The specific location found (determined) is stored in the database and used for various purposes. As a result, the convenience of the search service is improved.

2. Modifications The above embodiment can be variously modified. Specific modifications are exemplified below. Two or more aspects arbitrarily selected from the above embodiments and the following examples can be appropriately combined as long as they do not contradict each other.

2 (1). Modification 1
In the above embodiment, as specific action, it is illustrated that a user gets lost in a specific place. The specific action may be another action. For example, the specific action may be that the user searches for an electronic coupon at a specific location, or that the user searches for a parking lot at the specific location.

  When searching for an electronic coupon at a specific location is a specific action, by referring to the specific location table ST, a URL that provides the electronic coupon when the specific location is searched for on the web is displayed at the top of the search result. Is preferred. In addition, when searching for a parking lot at a specific location is a specific action, when a specific location is searched on the web or searched by the car navigation system, information on nearby parking lots and the availability of parking lots are provided. It is preferable.

2 (2). Modification 2
Each search log may include a positioning error (for example, a GPS positioning error and a base station positioning error) for the search execution location. In the above configuration, it is preferable that the candidate extraction unit 200 excludes a search log in which a positioning error exceeds an error threshold from a target for extracting a candidate for a specific location.

2 (3). Modification 3
In the above embodiment (step S1020), as the number of times the selection URL including the character string related to the specific action is selected, the frequency of the specific action being performed in the specific place candidate is higher. Specific behavior indicators are calculated. In step S1020 of this modification, a specific action index is calculated based on the degree of association with the specific action calculated in advance for each URL (for each website). More specifically, it is as follows.

  As shown in FIG. 8, the specific location discovery server DS of the present modification further includes a relevance level table RT that stores the relevance level between the website specified by the URL and the specific action for each URL. The degree of association between each URL and the specific action is calculated by executing machine learning based on a phrase included in the website specified by the URL. Note that the association degree table RT may be provided in another server device (for example, the search server SS).

  If the stray spot is exemplified and described as the specific place, the index calculation unit 210 relates the selection URL included in the plurality of search logs corresponding to one candidate for each of the stray spot candidates extracted by the candidate extraction unit 200. Combine the degrees to calculate the lost index. The above-described lost index indicates that the greater the value (total value), the higher the frequency of getting lost at the lost spot.

  Generally speaking, the index calculation unit 210 of this modification example increases the relevance value of the selected URLs included in a plurality of search logs including the location names corresponding to the candidates for each specific location candidate. The specific action index is calculated so as to indicate that the specific action is frequently performed in the candidate.

2 (4). Modification 4
In step S1020 of this variation, a specific action index is calculated based on the co-occurrence word in the search process. More specifically, it is as follows.

  The search log of this modification includes the first search word that is the place name and the second search word that is a co-occurrence word (word that is input simultaneously) with the first search word, which has already been described.

  When the stray spot is exemplified and described as the specific place, the index calculation unit 210 is lost in each of the plurality of search logs corresponding to the stray spot candidate for each of the stray spot candidates extracted by the candidate extraction unit 200. A search log that includes a second search term (eg, “access”, “traffic”, “how to”, “directions”, “where”, etc.) Count and calculate the lost index. The above-described ambiguity index indicates that the greater the value, the higher the frequency of ambiguity at the ambiguity spot.

  In other words, the index calculation unit 210 of the present modified example has a higher frequency of performing a specific action on each candidate at a specific place as the number of search logs including the second search word related to the specific action increases. Calculate specific behavior indicators to show that they are high.

  Note that the information on the selected URL is not essential in the calculation of the specific behavior index of the present modification. Therefore, the search log transmission unit 120 of the search server SS may not transmit the information on the selected URL to the specific location discovery server DS.

2 (5). Modification 5
There are cases where many people gather in one place (for example, a concert by a popular artist). In the above case, it is assumed that many people search for a nearby location at the same time (for example, searching for the nearest station after a concert). In the above case, there is a possibility that the location in the vicinity is determined to be a specific location although it is not originally a specific location.

  Therefore, in this modified example, it is determined whether or not the specific place is in consideration of the period during which the search process is performed. More specifically, the time is divided into time slots of a predetermined period (for example, 10 minutes / 1 time slot), a specific place is determined for each time slot, and the specific place is determined based on a plurality of determination results. Determine whether or not. For example, if a candidate for a specific location has a number of times (time slots) determined to be a specific location more than 10,000 times in one year, the candidate is determined to be a specific location.

  That is, in this modification, the index calculation unit 210 calculates a specific action index in each of a plurality of periods (time slots). The specific place determination unit 220 determines whether or not the candidate is a specific place based on a plurality of specific action indexes calculated over a plurality of periods.

  In addition, in order to further improve the determination accuracy, the determination process may be performed with outliers excluded. For example, when a specific action index in a certain time slot exceeds N times (for example, 10 times) a statistical value (average value, median, mode, etc.) of the specific action index in a plurality of time slots, the specific action A configuration that is not used for the determination process by the specific location determination unit 220 may be employed assuming that the index is an abnormal value.

2 (6). Modification 6
In the above embodiment, the specific location discovery server DS which is one device executes the event information collection operation. However, the specific location discovery system constituted by a plurality of devices may execute the event information collection operation. For example, the candidate extraction unit 200, the index calculation unit 210, the specific location determination unit 220, the search log table LT, and the specific location table ST may be implemented by an information management system including two or more server devices.

2 (7). Modification 7
Each function executed by the CPU in the computer apparatus, particularly the search server SS and the specific location discovery server DS, may be executed by a programmable logic device such as an FPGA (Field Programmable Gate Array) or a DSP (Digital Signal Processor).

DESCRIPTION OF SYMBOLS 10 ... Network interface, 12 ... Input part, 14 ... Output part, 16 ... CPU, 18 ... RAM, 20 ... ROM, 22 ... HDD, 30 ... Wireless communication interface, 32 ... Input device 34 …… Output device, 36 …… GPS module, 38 …… CPU, 40 …… RAM, 42 …… ROM, 44 …… Flash memory, 100 …… Search processing unit, 110 …… Search result presentation unit, 120 …… Search log transmission unit, 200 …… Candidate extraction unit, 210 …… Index calculation unit, 220 …… Specific location determination unit, DS …… Specific location discovery server, LT …… Search log table, RN …… Wireless communication network , SS ... search server, ST ... specific location table, UE ... user equipment.

Claims (9)

A specific action is performed from a plurality of search logs based on a search execution place included in the search log where the user has executed a search process and a place name which is a search term input by the user in the search process. A candidate extraction unit that extracts candidates for a specific place that is an easy place;
An index calculation unit that calculates a specific action index corresponding to the frequency with which the specific action is performed for each candidate of the extracted specific place;
A specific location determination unit that determines whether the candidate is the specific location based on the calculated specific behavior index, and
The search log
A selection URL selected by the user from a search result URL that is an output of the search process,
The indicator calculation unit
A specific place finding device that calculates the specific action index based on the selected URL for each candidate of the specific place. The candidate extraction unit
The location name is extracted as a candidate for the specific location when a distance between the search execution location included in one search log and a location indicated by the location name included in the search log is less than a threshold. Specific location discovery device. The indicator calculation unit
For each candidate at the specific location, the more frequently the selected URL including the character string related to the specific action is selected, the more frequently the specific action is performed on the candidate corresponding to the selected URL. The specific place finding device according to claim 1 or 2 , wherein the specific action index is calculated as shown. For each URL, including a relevance table that stores the relevance between the website specified by the URL and the specific action,
The indicator calculation unit
For each candidate of the specific location, the greater the sum of the relevance values of the selected URLs included in the plurality of search logs including the location name corresponding to the candidate, the more frequently the specific action is performed on the candidate. said specific location finding device according to claim 1 or 2 calculates a specific action index to indicate that high. The search log
A first search word that is the location name and a second search word that is a co-occurrence word of the first search word;
The indicator calculation unit
For each candidate at the specific location, the specific action index is set to indicate that the more the search log including the second search word related to the specific action is, the higher the frequency at which the specific action is performed in the candidate. The specific location finding device according to claim 1 or 2. The specific action is
Wherein the said user at a particular location Lost, wherein in certain locations that the user searches the electronic coupon, and wherein said user in a particular location one of claims 1 is either searching a parking 5 Specific location discovery device. The search log
Including a positioning error for the search execution location,
The candidate extraction unit
The specific location discovery apparatus according to any one of claims 1 to 6 , wherein a search log in which the positioning error exceeds an error threshold is excluded from a target from which the specific location candidate is extracted. The indicator calculation unit
Calculating the specific behavior index in each of a plurality of periods;
The specific place determination unit
The specific location discovery apparatus according to any one of claims 1 to 6 , wherein it is determined whether or not the candidate is the specific location based on the plurality of specific behavior indexes calculated over a plurality of the periods. From multiple search logs, included in the search log, and search execution where the user has performed a search process, the based on the location name is a search word input by the user in the search process, the specific action line extracting a candidate for a particular location is a fragile place,
For each candidate of the extracted said specific location, a step of calculating a specific action index corresponding to the frequency of the specific action is performed,
Based on the calculated the specific action index, the candidate is a step of determining whether or not the specific location,
The search log
A selection URL selected by the user from a search result URL that is an output of the search process,
In the step of calculating the specific behavior index,
For each candidate for the specific location, calculate the specific action index based on the selected URL
A place finding program that lets a computer do things .

Images