JP7066958B2 - Visit recommendation device, visit recommendation method and visit recommendation program - Google Patents

Description

The present invention relates to a visit destination recommendation device, a visit destination recommendation method, and a visit destination recommendation program that recommends a visit destination to a moving body such as a visitor from transit information in an arbitrary area.

Various methods have been proposed to recommend destinations to moving objects such as visitors based on information such as population distribution (people flow pattern) in a given area. People flow patterns are also called transit information.

Non-Patent Document 1 discloses a technical idea of analyzing migratory behavior between areas in units of 250 m (meters) mesh by utilizing mobile spatial statistics.

Non-Patent Document 2 discloses a technical idea of probabilistically estimating a stop-by place by a hierarchical migration model based on questionnaire information.

Patent Document 1 discloses a "product information providing system" that supports the purchase of products in a physical store. In Patent Document 1, by analyzing and estimating the flow line from the camera or GPS (Global Positioning System) information, the interest of the visitor is grasped and the recommended product is determined.

Patent Document 2 discloses a "human flow analysis system" that supports efficient operation of a city. In Patent Document 2, the start point and the end point are displayed on the map, and the waypoints are further displayed. Commercial operators, developers and transportation operators can compare wide-area migration patterns that pass through different stopovers. The end point is, for example, a predetermined facility. In Patent Document 2, it is possible to compare the admission rates for each facility and further grasp the migration pattern of people using the predetermined facility. The excursion pattern extracted by the excursion pattern extraction process can be used by transportation companies for transportation planning, by commercial companies for tourism planning and marketing, and by developers as hints for promotion planning. Patent Document 2 can also be used for navigation of tourist bases and recommendation of tour courses and destinations.

Patent Document 3 discloses an "information providing system" that can predict a user's behavior pattern as accurately as possible and select and provide the optimum information to the user at each time point. Patent Document 3 describes a method of determining the degree of interest for each genre for each unit section from the viewpoint of "what genre each user visited and used the store to which the individual user belongs". .. In Patent Document 3, the stay coefficient calculated for each common neighborhood store is aggregated as a value indicating the degree of interest in the store genre of the common neighborhood store. In Patent Document 3, such aggregation is performed for a large number of users over a predetermined period, and a visiting usage interest level table is created with the aggregated value as the degree of interest.

In Patent Document 3, it is necessary to investigate the behavior of a large number of users and perform statistical analysis in order to obtain the degree of interest under specific conditions. Therefore, in Patent Document 3, a questionnaire survey is conducted for a large number of users in a specific time zone.

Further, Patent Document 4 discloses a technical idea for presenting content information suitable for a user. The information processing apparatus disclosed in Patent Document 4 includes a preference / situation consideration content evaluation unit. The Preference / Situation Consideration Content Evaluation Unit calculates a synthetic similarity that combines the preference similarity by distance and the situation similarity. The synthetic similarity is the similarity between the requesting user of the recommendation request and the user who has selected the content information (search content information) in the past, and corresponds to the user value similarity. In Patent Document 4, it is possible to recommend a candidate point suitable for the user's preference and the user's current situation even in an area far from the user's activity area.

Japanese Unexamined Patent Publication No. 2015-022354 International Publication No. 2016/067369 International Publication No. 2009/075043 JP-A-2015-056045

Analysis of pedestrian migration in the city center (Part 2) -Attributes of residents in the city center-, Nagoya City Center Research Report, No. 117 2015. Predictive model for detailed analysis and policy simulation of migratory behavior in Machinaka, JSCE Proceedings, Vol. 71, No.5 323-335, 2015

Each of the above-mentioned prior arts (Patent Documents 1 to 4 and Non-Patent Documents 1 and 2) has the following problems.

Since Patent Document 1 utilizes information from a camera and GPS, it is possible to acquire an individual's action trajectory. However, since the technique of Patent Document 1 leads to individual identification, there is a high possibility of infringing privacy.

Since Non-Patent Document 1 uses population distribution information such as mobile spatial statistics, it is possible to protect privacy. However, in the method disclosed in Non-Patent Document 1, it is difficult to estimate fine migratory behavior because the analysis particle size is rough.

In the method disclosed in Non-Patent Document 2, migratory behavior is analyzed based on population distribution and questionnaire information based on a probabilistic behavior model. However, with such an analysis method, it is not possible to grasp the detailed relationships between facilities. Therefore, the method disclosed in Non-Patent Document 2 is insufficient information for recommending a visitor to a visitor based on the analysis result.

Patent Document 2 merely discloses a technical idea that makes it possible for a commercial business operator, a developer, or a transportation business operator to understand a person's migration pattern. Therefore, Patent Document 2 does not appropriately recommend a visitor to a visitor based on the migration pattern.

Patent Document 3 also discloses a technical idea of selecting and providing optimal information to a user. In order to realize this, in Patent Document 3, it is necessary to conduct a questionnaire survey on a large number of users.

Patent Document 4 merely discloses the technical idea of recommending a candidate point suitable for the user's preference and the user's current situation even in an area far from the user's activity area. .. Even in Patent Document 4, since the relationship between facilities is not grasped, it is not recommended to the visitor an appropriate destination.

One of the objects of the present invention is a visit destination recommendation device, a visit destination recommendation method, and a visit that can solve the above-mentioned problems and recommend an appropriate visit destination to a moving body such as a visitor who behaves in a similar manner. It is to provide a pre-recommendation program.

As one aspect of the present invention, when the visit recommendation device passes through the first facility in the area where the given moving body is located, based on the transit information including a plurality of facilities that the moving body has traveled along the route. With respect to the first route of the above, a selection means for selecting a facility in which the given moving body does not pass through the first route in the route that passes through the first facility in the route information; It is provided with a determination means for determining a second route via one facility and the facility selected as a destination based on map information indicating a road map of the certain area; the moving body is a person, and the moving body is a person. The transit information consists of a traffic pattern that is a movement in a traffic group unit in a certain area, each of the plurality of facilities contains a store of at least one industry, and the selection means is a plurality of. An interest level calculation means for calculating the degree of interest in an industry for an industry included in a facility in the area based on a person flow pattern and facility information indicating a plurality of facilities in the area; based on the degree of interest in the industry. A similarity calculation means for calculating the inter-group similarity and the inter-facility similarity of the human flow pattern; the determination means is based on the map information, the inter-group similarity, and the inter-facility similarity. , A travel recommendation determination means for recommending a travel route is provided so that a specific facility having a store in an industry not included in a given flow pattern is the destination of the visit .

As another aspect of the present invention, the visit recommendation method is a first method in an area where a given moving body is located, based on transit information including a plurality of facilities that the moving body has traveled along the route traveled by the information processing device. With respect to the first route when passing through the facility, the facility in which the given moving body does not pass through the first route in the route passing through the first facility in the transit information is selected; The second route via the first facility and the selected facility as a visit destination is determined based on the map information showing the road map of the certain area, the moving body is a person, and the transit information is It consists of a person flow pattern, which is a movement in units of a person flow group in the certain area, and each of the plurality of facilities contains a store of at least one industry, and the plurality of people flow patterns and a plurality of the above-mentioned areas. Based on the facility information indicating the facility, the degree of industry interest in the industry included in the facility in the certain area is calculated; based on the degree of industry interest, the degree of similarity between groups and the degree of similarity between the facilities of the person flow pattern Based on the map information, the similarity between groups, and the similarity between facilities, the visit destination is a specific facility having stores in industries not included in a given flow pattern. As such, we recommend a tour route .

In another aspect of the present invention, the visit recommendation program is based on transit information including a plurality of facilities traversed by the route traveled by the mobile vehicle when the mobile vehicle traverses a first facility in an area. With respect to the first route of the above, a selection procedure for selecting a facility in which the given moving body does not pass through the first route in the route that passes through the first facility in the transit information; A computer is made to execute a decision procedure for determining a second route via one facility and the facility selected as a destination based on map information indicating a road map of a certain area, and the moving body is a person. Therefore, the via information consists of a person flow pattern that is a movement in a person flow group unit in the certain area, and each of the plurality of facilities contains a store of at least one industry, and the selection procedure is described. An interest degree calculation procedure for calculating an industry interest level for an industry included in a facility in a certain area based on a plurality of people flow patterns and facility information indicating a plurality of facilities in the certain area on the computer; A similarity calculation procedure for calculating the inter-group similarity and the inter-facility similarity of the person flow pattern based on the degree of interest in the industry; A tour recommendation determination procedure that recommends a tour route so that a specific facility having a store in an industry not included in a given flow pattern is the destination of the visit based on the similarity and the similarity between the facilities. To execute .

Further, it is realized by a recording medium that can be read by a computer that records such a program.

According to the present invention, it is possible to recommend an appropriate destination for a moving body such as a visitor who behaves in a similar manner.

It is a block diagram which shows the structure of the visit destination recommendation system which includes the visit destination recommendation apparatus which concerns on one Embodiment of this invention. It is a block diagram which shows the main processing part realized by the visit destination recommendation apparatus used in the visit destination recommendation system shown in FIG. 1. It is a flowchart for demonstrating the schematic operation of the visit destination recommendation apparatus shown in FIG. It is a figure which shows an example of the pedestrian map of a city. It is a flowchart for demonstrating the operation of the behavior analysis unit used in the visit recommendation apparatus shown in FIG. It is a figure which shows an example of the information of the person flow pattern stored as the processing information in the storage device used for the visitor recommendation system shown in FIG. 1. It is a figure which shows an example of facility information stored as processing information in the storage device used for the visit recommender system shown in FIG. 1. It is a figure which shows the relationship between the person flow pattern and the industry interest degree Bint calculated for each industry type B type. It is a figure which showed the inter-industry similarity score Bscore _{i, j between} the th-th industry type B type _i and the j-th industry type B type j in a table. It is a figure which showed the person flow pattern similarity score PF score _{i, j} between the th-th person flow pattern PF _{i, T} and the th-th person flow pattern PF _{j, T} in a table. It is a block diagram which shows the relationship between the measure decision part and the person flow simulator used in the visit recommender system shown in FIG. It is a flowchart for demonstrating the operation of the measure decision part used in the visit destination recommendation apparatus shown in FIG. It is a figure for demonstrating an example in which a migration recommendation determination means recommends an industry from the similarity between facilities (similarity of an industry / type). It is a figure for demonstrating an example in which a migration recommendation determination means recommends an industry from the similarity between groups (similarity between people flow patterns). It is a figure for demonstrating an example of the method which the migration recommendation determination means determines the place of the migration recommendation.

[Embodiment]
FIG. 1 is a block diagram showing a configuration of a visitor recommendation system 100 including a visitor recommendation device (described later) according to an embodiment of the present invention.

The illustrated visit recommendation system 100 outputs a data processing device (information processing device) 200 for processing data, a storage device 300 for storing programs and data described later, an input device 400 for inputting data, and data. It includes an output device 500 and.

The output device 500 includes a display device such as an LCD (Liquid Crystal Display) or a PDP (Plasma Display Panel) or a printer. The output device 500 has a function of displaying various information such as an operation menu and printing out the final result in response to an instruction from the data processing device 200.

The storage device 300 includes a memory such as a hard disk, a read-only memory (ROM), and a random access memory (RAM). The storage device 300 has a function of storing the processing information 310 and the program 320 required for various processes in the data processing device 200.

The data processing device 200 includes a microprocessor such as an MPU (micro processing unit) and a central processing unit (CPU). The data processing device 200 has a function of reading the program 320 from the storage device 300 and realizing various processing units that process data according to the program 320.

The data processing device 200 includes a human flow simulator 210 and a visit destination recommendation device 220.

In the illustrated visitor recommendation system 100, data required for an algorithm, a flow pattern at a region and time at which a visitor is desired to be recommended, such as device mobile spatial statistics, map information, and facility information data are given. It is assumed that there is.

Here, the people flow pattern is a pattern showing movement in units of people flow groups in the city. This human flow pattern may be given by observing (measuring) in advance, or may be estimated by a human flow pattern estimation system (not shown). In this example, the human flow pattern is not the pattern itself, but a parameter supplied to the human flow simulator 210. The parameter produces data from the human flow simulator 210 that is closest to the population distribution (human flow pattern) in a given area when it is supplied to the human flow simulator 210. Specifically, the "parameter" is the "number of people in the group, where they came from (departure point), what time they came from (departure time), where they returned from (arrival point), and where they went through. It consists of "whether to do it (route), the staying time of the transit value". The starting point and the reaching point may be the same.

The map information indicates a road map in the city.

Facility information indicates multiple facilities in the city. It should be noted that each of the plurality of facilities contains at least one type of business. The industries include, for example, the restaurant industry, the clothing industry, the retail industry, and the medical industry. The restaurant industry includes, for example, sushi, Chinese food, and taverns. The clothing industry includes, for example, women's clothing and children's clothing. The retail industry includes, for example, convenience stores and supermarkets. The medical industry includes, for example, internal medicine and plastic surgery.

These people flow patterns, map information, facility information, and the like are stored in advance in the storage device 300 as processing information 310.

The visit destination recommendation device 220 of the visit destination recommendation system 100 is a device that recommends a visit destination to a visitor based on a person flow pattern, map information, and facility information. In the embodiment described below, a case where an appropriate visit destination is recommended mainly to a person (visitor) who behaves similarly is described as an example, but the present invention is described for a person (visitor). It is not limited, and is generally applicable to moving objects. In this case, the human flow pattern can be replaced with transit information.

FIG. 2 is a block diagram showing a main processing unit realized by the visit destination recommendation device 220. The visit recommendation device 220 includes a behavior analysis unit 600 and a policy decision unit 700.

The schematic operation of the visit recommendation device 220 will be described with reference to FIG.

First, the behavior analysis unit 600 calculates the relationship information of interests between facilities based on a plurality of people flow patterns and facility information (step S101).

Subsequently, the policy-making department 700 visits a specific facility with an industry that has not been visited before for a specific group of people, based on the map information and the information on the relationship of interests between the above facilities. The migration route is recommended (step 102). In other words, the policy-making unit 700 recommends a migration route so that the visit destination is a specific facility having an industry that is not included in the given flow pattern.

Returning to FIG. 2, the configuration of the visited destination recommendation device 220 will be described in more detail.

The behavior analysis unit 600 includes an interest degree calculation means 610 and a similarity calculation means 620. The measure determination unit 700 includes a migration recommendation determination means 710.

The interest degree calculation means 610 calculates the industry interest degree for the industry included in the facility in the city based on a plurality of people flow patterns and facility information. That is, the interest degree calculation means 610 calculates the industry interest degree indicating which industry is interested in and how much from the movement history of each group of the people flow pattern and the industry included in the facility to stop by.

The similarity calculation means 620 calculates the similarity between groups and the similarity between facilities of the flow pattern based on the degree of interest in the industry. The similarity between groups and the similarity between facilities are stored as processing information 310 in the storage device 300.

The migration recommendation determination means 710 recommends the migration route based on the map information, the similarity between groups, and the similarity between facilities. That is, the migration recommendation determination means 710 visits a place of interest that has not been stopped by a group of people who behave similarly based on the degree of similarity between groups and the degree of similarity between facilities. Recommended as a destination. The migration recommendation information indicating the recommended migration route is stored in the storage device 300 as processing information 310.

Therefore, the information on the relevance of interests between the facilities consists of a combination of the degree of interest in the industry, the degree of similarity between groups, and the degree of similarity between facilities.

In addition, the migration recommendation determination means 710 provides information on the joint campaign between facilities having a high degree of similarity based on the degree of interest in the industry and the degree of similarity between facilities. Specifically, for example, in the clothing industry, it is assumed that women's clothing stores and children's clothing stores have a high degree of similarity. In this case, the migration recommendation determination means 710 implements a campaign such as issuing a discount ticket to the other store (set sale) or giving preferential treatment to the person who purchased at one store.

Next, the operation of the visit recommendation device 220 will be described in more detail.

FIG. 4 is a diagram showing an example of a pedestrian map of the city. The illustrated pedestrian map is obtained from map information and a flow pattern.

In the illustrated example, as a plurality of people flow patterns, an example of two people flow groups, a first person flow pattern PF _{1 and T} and a second person flow pattern PF _{2 and T} , is given. Further, on the map, two starting points, a _first starting point S1 and a _second starting point S2, are drawn as starting points. In the illustrated example, the first departure point S1 is also the _first arrival point, and the second departure point S2 is also the _second arrival point.

Further, on the map, the first to fifth facilities P ₁ , P ₂ , P ₃ , P ₄ , P ₅ and the nth facility P _n are drawn as a plurality of facilities. In the illustrated example, the _first facility P1 contains a Chinese restaurant, a tavern, and a children's clothing store as stores. The _second facility, P2, contains sushi and izakaya stores as stores. The _third facility, P3, contains internal medicine and plastic surgery stores as stores. The _fourth facility, P4, contains stores for women's clothing, children's clothing, convenience stores, and supermarkets. The _fifth facility, P5, contains a women's clothing store as a store. The _nth facility Pn contains stores for women's clothing and children's clothing as stores.

The first people flow patterns PF _{1 and T} , which are the first people flow groups, have a _second starting point S2 as a starting point and a _second facility P2 and a _fourth facility P4 as transit points. It is a human flow pattern in which the arrival point _S2 of is the arrival point.

The second flow patterns PF _{2 and T} , which are the second flow groups, have the first starting point S ₁ as the starting point, and the first facility P ₁ , the third facility P ₃ , and the fifth facility P. This is a human flow pattern in which ₅ is a stopover and the _first arrival point S1 is the arrival point.

The operation of the behavior analysis unit 600 will be described with reference to FIG.

First, the interest degree calculation means 610 calculates the industry interest degree for the industry included in the facility in the city based on the person flow pattern PF and the facility Ftype represented by the facility information (step S201). A specific example of the degree of interest in the industry will be described later with reference to the drawings.

Subsequently, the similarity calculation means 620 calculates the similarity between groups of the flow pattern from the degree of interest in the industry (step S202). A specific example of the similarity between groups will also be described later with reference to the drawings.

Finally, the similarity calculation means 620 calculates the similarity between facilities from the degree of interest in the industry (step S203). The similarity between facilities will also be described later with reference to the drawings.

FIG. 6 is a diagram showing an example of information on a human flow pattern stored as processing information 310 in the storage device 300.

FIG. 6 shows an example in which there are first to nth human flow patterns PF _{1, T} to PF _{n, and T} as a plurality of human flow patterns. Generally, in the i-th human flow pattern PF _{i, T} (1 ≦ i ≦ n), the subscript i indicates the pattern identifier, and the subscript T represents the time zone {am / pm}. It is assumed that each person flow pattern goes through any one of the _first to _nth facilities P1 to Pn as a waypoint. Here, in the facility Pi of the _i -th, the subscript i indicates a pattern identifier.

In the example shown in FIG. 6, for example, the first human flow patterns PF _{1 and T} are a human flow group consisting of 100 people, and pass through the _second facility P2 and the _fourth facility P4 as transit points. Shows that to do. Further, the second flow patterns PF _{2 and T} are a flow group consisting of 200 people, and pass through the first facility P ₁ , the third facility P ₃ , and the fifth facility P ₅ as transit points. Shows that to do.

FIG. 7 is a diagram showing an example of facility information stored as processing information 310 in the storage device 300.

As shown in FIG. 7, the facility information includes _first to _nth facilities P1 to Pn, and industry information indicating the type of industry included in each facility. Industry information is identified by industry type Btype. The presence or absence of stores in each facility is indicated by "1" and "0", respectively.

In the example shown in FIG. 7, as facility information, the _first facility P1 contains a Chinese restaurant and a pub store as an industry (store), and a children's clothing store as a clothing business. Is shown. Further, it is shown that the _second facility P2 contains a restaurant business sushi and a pub store as an industry (store). The _third facility, P3, indicates that the medical industry's internal medicine and plastic surgery stores are included as industries (stores). The _fourth facility, P4, has stores for women's clothing and children's clothing in the clothing industry as industries (stores), and stores for convenience stores and supermarkets in the retail industry. It is shown that the _fifth facility P5 contains a women's clothing store in the clothing industry as an industry (store). Then, it is shown that the _nth store Pn includes stores for women's clothing and children's clothing in the clothing industry as an industry (store).

In addition, as facility information of the building that is the _fourth facility P4, there is an example where a convenience store and a supermarket store are on the first floor, a women's clothing store is on the second floor, and a children's clothing store is on the fourth floor. Is shown.

The pedestrian map shown in FIG. 4 is a diagram created from, for example, the information on the flow pattern shown in FIG. 6 and the facility information shown in FIG. 7.

With reference to FIG. 8, the industry interest degree Bint calculated by the interest degree calculation means 610 will be described.

FIG. 8 is a diagram showing the relationship between the flow pattern and the industry interest level Bint calculated for each industry type B type.

As is clear from FIG. 8, the industry interest degree Bint is calculated from the information of the person flow pattern PF shown in FIG. 6 and the facility information Ftype shown in FIG. 7 according to the following number 1.

For example, the industry interest level Bint of the third people flow patterns PF _{3 and T} , which is the third people flow group, is calculated as follows. From FIGS. 4 and 6, the third human flow patterns PF _{3 and T} start from the _first starting point S1 and pass through the _first facility P1 and the _second facility P2 to the second. It is a flow pattern consisting of 150 people, with the arrival point _S2 as the arrival point. As shown in FIG. 7, the _first facility P1 contains stores for Chinese food, taverns, and children's clothing, and the _second facility P2 contains stores for sushi and taverns. Therefore, the industry interest level Bint of the third person flow pattern PF _{3 and T} is equal to 150 for sushi, equal to 150 for Chinese food, and 300 for izakaya (that is, 150 for izakaya in facility P ₁ ) for each industry type B type. And 150 for _taverns in facility P2), and children's clothing is equal to 150.

ここで、Ｎは人流パターンＰＦの人数を示し、ｐ_ｋは第ｋの人流パターンＰＦ_ｋ、Ｔの第ｉの業種タイプＢtype_ｉの業種興味度を示し、ｑ_ｋは第ｋの人流パターンＰＦ_ｋ、Ｔの第ｊの業種タイプＢtype_ｊの業種興味度を示す。 Next, the inter-facility similarity calculated by the similarity calculation means 620 will be described. The inter-facility similarity is composed of the inter-industry similarity score B score represented by the following number 2.

Here, N indicates the number of people in the flow pattern PF, p _k indicates the kth person flow pattern PF _{k, T} indicates the degree of industry interest in the _i -th industry type B type i, and q _k indicates the k-th person flow pattern PF _k . _{, T} indicates the degree of industry interest of the _jth industry type B type j.

FIG. 9 is a diagram showing the inter-industry similarity scores Bscore _{i and j between} the i-th industry type B type _i and the j-th industry type B type j calculated by the above equation 2 in a table. The inter-industry similarity scores B score _{i and j} are stored in the area of the processing information 310 of the storage device 300 as the inter-facility similarity (see FIG. 2).

ここで、Ｎは業種タイプＢtypeの数を示し、ｐ_ｋは第ｋの業種タイプＢtype_ｋの第ｉの人流パターンＰＦ_ｉ，Ｔの業種興味度を示し、ｑ_ｋは第ｋの業種タイプＢtype_ｋの第ｊの人流パターンＰＦ_ｊ，Ｔの業種興味度を示す。 Next, the intergroup similarity of the human flow pattern calculated by the similarity calculation means 620 will be described. The similarity between groups of the flow pattern consists of the PF score of the similarity pattern of the flow pattern, which is represented by the following number 3.

Here, N indicates the number of industry type B type, p _k indicates the degree of industry interest of the i-th person flow pattern PF _{i, T} of the _kth industry type B type k, and q _k indicates the industry interest level of the kth industry type B type _k . The jth person flow pattern PF _{j, T} shows the degree of interest in the industry.

FIG. 10 tabulates the human flow pattern similarity score PF score _{i, j} between the i-th human flow pattern PF _{i, T} and the j-th human flow pattern PF _{j, T} calculated by the above equation 3. It is a figure. The human flow pattern similarity scores _{i and j} are stored in the area of the processing information 310 of the storage device 300 as the similarity between groups (see FIG. 2).

In other words, the person flow pattern similarity score indicates the degree to which the two people flow patterns are similar from the viewpoint of the store of the type of business visited by a person. For example, the larger the similarity score, the higher the degree of similarity between the two flow patterns. Conversely, the smaller the similarity score, the lower the degree to which the two patterns of human flow are similar. That is, by selecting two people flow patterns with high similarity scores, it is possible to select similar people flow patterns from the viewpoint of stores in the type of industry visited.

From the above explanation, the behavior analysis unit 600 is based on the transit information (for example, a human flow pattern) including a plurality of facilities that the mobile body (for example, a person) has traveled along the route, and the area where the given mobile body is located. As a means of selecting a facility in which a given moving object does not pass through the first route in the route passing through the first facility in the route information regarding the first route when passing through the first facility in the above. work.

Next, the operation of the measure determination unit 700 (migratory recommendation determination means 710) will be described with reference to FIGS. 11 and 12. FIG. 11 is a block diagram showing the relationship between the measure decision unit 700 and the human flow simulator 210. FIG. 12 is a flowchart for explaining the operation of the measure determination unit 700 (migratory recommendation determination means 710).

Referring to FIG. 12, first, the migration recommendation determination means 710 selects at least one migration pattern PF _{i, T} having a low migration rate as a specific migration group based on the map information (step S301).

Subsequently, the migration recommendation determination means 710 selects the top M recommended industries recommended for the specific human flow groups PF _{i, T} based on the inter-facility similarity shown in FIG. 9 and the inter-group similarity shown in FIG. Select (step S302). Here, the recommended industry to be recommended is an industry with a high (similarity score x number of people).

Next, the detour recommendation determination means 710 selects a plurality of specific facility Pis having the selected recommended industry, and a specific person flow group PF _{i ,} so as to pass through each of the selected specific facility _Pis . A plurality of detour route candidates are added for _T and evaluated by the human flow simulator 210 (step S303).

More specifically, as shown in FIG. 11, the migration recommendation determination means 710 selects a specific facility _Pi to be added to the migration route and supplies it to the human flow simulator 210. The human flow simulator 210 simulates a migration route to which the specific facility _Pi is added, and returns positive rewards and negative rewards to the policy determination unit 700 (migration recommendation determination means 710). Here, the positive reward is output when there is a recommendation effect. On the other hand, the negative reward is output when the moving distance of the human flow patterns PF _{i and T} increases.

Returning to FIG. 12, the migration recommendation determination means 710 determines P industries / facilities having a large sum of positive rewards and negative rewards from the plurality of migration route candidates (step S304). In other words, the migration recommendation determination means 710 determines one recommended migration route from the plurality of migration route candidates so that at least one specific facility having a high-paying industry is the destination of the visit.

Next, with reference to FIGS. 13 and 14, an example will be described in which the migration recommendation determination means 710 determines the migration recommendation industry when the migration recommendation is made to the specific human flow groups PF _{i, T.}

FIG. 13 is a diagram for explaining an example in which the migration recommendation determination means 710 recommends an industry based on the similarity between facilities (similarity of industry / type).

First, from FIG. 13, the migration recommendation determination means 710 is a clothing industry (children's clothing) as an industry type having a high degree of similarity to the current visit destination of a specific person flow group (in this example, the i person flow pattern) PF _{i, T.} ) And the medical industry (internal medicine, orthopedics). Here, since the medical industry (internal medicine, plastic surgery) is the same as the current visit destination, the migration recommendation determination means 710 adds the clothing industry (children's clothing) to the candidate list.

FIG. 14 is a diagram for explaining an example in which the migration recommendation determination means 710 recommends an industry based on the similarity between groups (similarity between people flow patterns).

First, from FIG. 14, the migration recommendation determination means 710 selects the second human flow patterns PF _{2 and T} as the human flow patterns having high similarity to the first human flow patterns PF _{i and T.} Subsequently, the wandering recommendation determination means 710 is the second person flow among the visit destinations of the second person flow patterns PF _{2 and T} based on the information of the person flow pattern stored in the storage device 300 (see FIG. 6). Add the restaurant business (Chinese, tavern) and clothing business (children's clothing) that the patterns PF _{i and T} have not visited to the candidate list. Here, among these industries (stores), the recommendation effect regarding the clothing industry (children's clothing) is 1662, which is the highest. Therefore, when M = 1, the traveling recommendation determination means 710 recommends the clothing industry (children's clothing) because the recommendation effect of the clothing industry (children's clothing) is the highest.

Next, with reference to FIG. 15, when the migration recommendation determination means 710 recommends the clothing industry (children's clothing), an example of a method of determining the location of the migration recommendation will be described.

FIG. 15 shows that when the i-th person flow pattern PF _{i, T} is selected as a specific person flow group and the clothing industry (children's clothing) is recommended, the detour place is recommended to the i-th person flow pattern PF _{i, T.} It is a figure which shows an example.

As is clear from FIG. 6, the current i-th person flow pattern PF _{i, T} is a person flow group consisting of 300 people, and passes through the _third facility P3 and the _fifth facility P5 as transit points. Shows that to do.

Referring to the “current PF _{i, T} ” in FIG. 15, the third person flow pattern PF _{i, T} , which is the third person flow group, has the _first starting point S1 as the starting point and the _third facility P3. It is a human flow pattern in which the _fifth facility P5 is used as a stopover and the _first arrival point S1 is the arrival point.

First, when the migration recommendation determination means 710 recommends the clothing industry (children's clothing) to the i-th person flow patterns PF _{i and T} , a plurality of facility candidates (in this example, the first facility P ₁ and the first). A plurality of migration route candidates are added so as to pass through each of the facility P4 of ₄ and the facility Pn) of the _nth facility.

In this example, the first migration route candidate has the _first starting point S1 as the starting point, the _first facility P1, the _third facility P3, and the _fifth facility P5 as transit points. It is a migratory route with the arrival point S1 of ₁ as the arrival point. The second migration route candidate has a _first starting point S1 as a starting point, a _third facility P3, a _fifth facility P5, and a _fourth facility P4 as transit points, and a second arrival point. It is a migratory route with S ₂ as the arrival point. The third migration route candidate has the _first starting point S1 as the starting point, the _third facility P3, the _fifth facility P5, and the _nth facility Pn as transit points, and the first arrival point. It is a migratory route with S ₁ as the arrival point.

Next, the migration recommendation determination means 710 acquires the travel distance evaluated by the human flow simulator 210 for the first to third migration route candidates. In the case of this example, as the first migration route candidate, the movement distance of "+ 100 m" is added to the third human flow patterns PF _{i and T.} As the second migration route candidate, the movement distance of "+ 50 m" is added to the third human flow pattern PFs _{i and T.} As for the third migration route candidate, the movement distance of "+ 500 m" is added to the third human flow pattern PFs _{i and T.}

Then, the migration recommendation determination means 710 recommends the migration place where the reward is maximized. In the case of this example, the sum of the positive and negative rewards is the maximum for the second migration route candidate that adds the _fourth facility P4 to the third human flow pattern PF _{i, T.} Therefore, the migration recommendation determination means 710 determines and recommends the second migration route candidate as one recommended migration route to the i-th person flow group.

In this way, the measure decision unit 700 functions as a decision means for determining the second route via the first facility and the facility selected as a visit destination based on the map information indicating the road map of the certain area. ..

As is clear from the above explanation, according to the present embodiment, it is possible to recommend an appropriate destination to a visitor who behaves in a similar manner. The reason is that when making a migratory recommendation to a human flow group, the recommendation effect is a positive reward, the physical distance is a negative reward, and the visit destination is recommended so that the expected value reward is maximized. Because.

It should be noted that the present invention is not limited to the above-described embodiment as it is, and at the implementation stage, the components can be modified and embodied within a range that does not deviate from the gist thereof. In addition, various inventions can be formed by appropriately combining a plurality of components.

Each part of the visit recommendation device may be realized by using a combination of hardware such as an information processing device and software. In the form of combining hardware and software, a visit recommendation program is deployed in RAM (random access memory), and the hardware of a processor such as a CPU (central processing unit) is operated based on the visit recommendation program. By doing so, each part is realized as various means. In addition, the visit recommendation program may be recorded on a recording medium and distributed. The visit recommendation program recorded on the recording medium is read into a memory via a wired, wireless, or recording medium itself, and operates a processor (CPU) or the like. Examples of recording media include optical disks, magnetic disks, semiconductor memory devices, hard disks, and the like.

To explain the above embodiment in another expression, the behavior analysis unit 600 (interest degree calculation means 610, similarity calculation means 620) is based on the visit destination recommendation program developed in the RAM, and the computer operating as the visit destination recommendation device is described. ) And the measure determination unit 700 (migratory recommendation determination means 710) can be operated as a combination.

Further, the specific configuration of the present invention is not limited to the above-described embodiment, and is included in the present invention even if there is a change within a range that does not deviate from the gist of the present invention.

Although the invention of the present application has been described above with reference to the embodiments, the invention of the present application is not limited to the above-described embodiment. Various changes that can be understood by those skilled in the art can be made within the scope of the present invention in terms of the configuration and details of the present invention.

100 Visit recommendation system 200 Data processing device 210 Human flow simulator 220 Visit recommendation device 300 Storage device 310 Processing information 320 Program 400 Input device 500 Output device 600 Behavior analysis unit 610 Interest level calculation method 620 Similarity calculation method 700 Measure decision unit 710 Travel recommendation decision method

Claims (7)

The first of the above-mentioned transit information regarding the first route when a given moving object has passed through the first facility in an area based on the transit information including a plurality of facilities that have been passed through the route traveled by the moving object. A selection means for selecting a facility in which the given moving body does not pass in the first route in the route via one facility, and a selection means.
A determination means for determining a second route via the first facility and the facility selected as a destination based on map information indicating a road map of a certain area.
In the visit recommendation device equipped with
The moving body is a person, and the transit information is composed of a person flow pattern that is a movement in a person flow group unit in the certain area, and each of the plurality of facilities contains a store of at least one industry. Ori,
The selection means is
An interest level calculation means for calculating the degree of interest in an industry for an industry included in a facility in a certain area based on a plurality of people flow patterns and facility information indicating a plurality of facilities in the certain area.
A similarity calculation means for calculating the similarity between groups and the similarity between facilities of the person flow pattern based on the degree of interest in the industry is provided.
Based on the map information, the similarity between groups, and the similarity between facilities, the decision-making means makes the visit destination a specific facility having stores in industries not included in a given migration pattern. A means of determining a migration recommendation that recommends a migration route is provided.
Visit recommendation device . The migration recommendation determination means provides information on a joint campaign between facilities having a high degree of similarity based on the degree of interest in the industry and the degree of similarity between the facilities.
The visit recommendation device according to claim 1 . The means for determining the migration recommendation is
As the specific person flow group, a person flow group having a low migration rate and a person flow pattern is selected.
At least one recommended industry to be recommended for the specific person flow group is selected based on the similarity between the groups and the similarity between the facilities.
Select a plurality of specific facilities having the selected recommended industry, and add candidate migration route candidates for the specific human flow group so as to pass through each of the selected specific facilities.
From the plurality of excursion route candidates, one recommended excursion route is determined so that one specific facility having a high-paying industry is the visit destination.
The destination recommendation device according to claim 1 or 2 . Based on the transit information including a plurality of facilities that the mobile body has traveled along the route traveled by the information processing device, the transit information regarding the first route when the mobile body has passed through the first facility in a certain area. Of the above, in the route that passes through the first facility, the facility that the given moving object does not pass through in the first route is selected.
The second route via the first facility and the selected facility as a destination is determined based on the map information showing the road map of the certain area.
In the method of recommending destinations
The moving body is a person, and the transit information is composed of a person flow pattern that is a movement in a person flow group unit in the certain area, and each of the plurality of facilities contains a store of at least one industry. Ori,
Based on the plurality of people flow patterns and facility information indicating a plurality of facilities in the certain area, the degree of industry interest in the industry included in the facilities in the certain area is calculated.
Based on the degree of interest in the industry, the degree of similarity between groups and the degree of similarity between facilities of the person flow pattern are calculated.
Based on the map information, the similarity between groups, and the similarity between facilities, the migration route is such that a specific facility having a store in an industry not included in a given flow pattern is the destination of the visit. To recommend,
How to recommend a destination . Conduct a joint campaign between facilities with high similarity based on the degree of interest in the industry and the similarity between facilities.
The visit recommendation method according to claim 4 . As the specific person flow group, a person flow group having a low migration rate and a person flow pattern is selected.
At least one recommended industry to be recommended for the specific person flow group is selected based on the similarity between the groups and the similarity between the facilities.
Multiple specific facilities with the selected recommended industry are selected, and multiple candidate migration routes are added for the specific human flow group so as to pass through each of the selected specific facilities.
From the plurality of candidate migration routes, one recommended migration route is determined so that one specific facility having a high-paying industry is the visit destination.
The visit recommendation method according to claim 4 or 5 . The first of the above-mentioned transit information regarding the first route when a given moving object has passed through the first facility in an area based on the transit information including a plurality of facilities that have been passed through the route traveled by the moving object. A selection procedure for selecting a facility in which the given moving object does not pass in the first route in the route via one facility, and a selection procedure.
A determination procedure for determining a second route via the first facility and the facility selected as a destination based on map information indicating a road map of a certain area, and
In a visit recommendation program that causes a computer to execute
The moving body is a person, and the transit information is composed of a person flow pattern that is a movement in a person flow group unit in the certain area, and each of the plurality of facilities contains a store of at least one industry. Ori,
The selection procedure is performed on the computer.
An interest level calculation procedure for calculating the degree of interest in an industry for an industry included in a facility in a certain area based on a plurality of people flow patterns and facility information indicating a plurality of facilities in the certain area.
Based on the industry interest level, the similarity calculation procedure for calculating the inter-group similarity and the inter-facility similarity of the person flow pattern is executed.
In the determination procedure, the computer is provided with a specific facility having stores in an industry that is not included in a given migration pattern, based on the map information, the similarity between groups, and the similarity between facilities. To execute the migration recommendation decision procedure that recommends the migration route so that it is a visit destination,
Visit recommendation program .

Images