JP2023032213A - Information processing device, information processing method, and information processing program - Google Patents

Abstract

To optimize search conditions to be proposed.SOLUTION: An information processing device according to the present application comprises: an action information acquisition unit which acquires action information on user actions taken for search results per search condition type; a calculation unit which calculates an influence degree at which selection and non-selection of a search condition type influence an action by using the action information acquired by the action information acquisition unit; and a proposal unit which proposes search conditions of a search condition type selected according to the influence degree calculated by the calculation unit to a user.SELECTED DRAWING: Figure 1

Description

There is an application for the application of Article 30, Paragraph 2 of the Patent Law https://realstate. Yahoo. co. jp/rent/search/ai/ Literature 1. pdf Publication date July 27, 2021

The present invention relates to an information processing device, an information processing method, and an information processing program.

Conventionally, for example, in a service that provides a search site, there is a search support system that proposes search conditions for narrowing down or expanding when there are many or few search results based on conditions specified by the user (see, for example, Patent Document 1). ).

JP 2008-152607 A

However, in the prior art, for example, there is room for improvement in terms of optimizing the proposed search conditions.

SUMMARY OF THE INVENTION It is an object of the present invention to provide an information processing apparatus, an information processing method, and an information processing program capable of optimizing proposed search conditions.

In order to solve the above-described problems and achieve the object, an information processing apparatus according to the present invention includes: a calculation unit for calculating, using the action information acquired by the action information acquisition unit, the degree of influence of the presence or absence of selection of the search condition type on the action, and the degree of influence calculated by the calculation unit a proposing unit for proposing to the user the search condition of the search condition type selected accordingly.

According to the present invention, the proposed search conditions can be optimized.

FIG. 1 is a diagram illustrating an example of information processing according to an embodiment. FIG. 2 is a block diagram illustrating a configuration example of the information processing apparatus according to the embodiment; 3 is a diagram illustrating an example of information stored in a real estate information storage unit according to the embodiment; FIG. 4 is a diagram illustrating an example of information stored in a click log storage unit according to the embodiment; FIG. 5 is a diagram illustrating an example of information stored in a user information storage unit according to the embodiment; FIG. 6 is a diagram illustrating an example of information stored in a user attribute storage unit according to the embodiment; FIG. 7 is a diagram illustrating an example of information stored in an impact storage unit according to the embodiment; FIG. FIG. 8 is a flowchart illustrating an example of influence degree calculation processing according to the embodiment. FIG. 9 is a flowchart illustrating an example of search condition proposal processing according to the embodiment. FIG. 10 is a hardware configuration diagram showing an example of a computer that implements the functions of the information processing apparatus according to the embodiment.

Embodiments for implementing an information processing apparatus, an information processing method, and an information processing program according to the present application (hereinafter referred to as "embodiments") will be described in detail below with reference to the drawings. The information processing apparatus, information processing method, and information processing program according to the present application are not limited to this embodiment.

[Embodiment]
[1. information processing]
First, an example of information processing according to the embodiment will be described with reference to FIG. FIG. 1 is a diagram illustrating an example of information processing according to an embodiment.

As shown in FIG. 1, the information processing device 10 according to the embodiment has, for example, a real estate information storage unit, and provides a user terminal 50, which is a client terminal, with various information related to real estate information. is. For example, the information processing device 10 provides various types of real estate information regarding real estate rental, sales, and the like. Note that the information processing device 10 is realized by, for example, a server device, a cloud system, or the like.

Note that real estate includes buildings and land, and the real estate information provided by the information processing apparatus 10 includes information on rental or sales of various real estate such as residences, stores, and land.

The user terminal 50 is a terminal device owned by the user U, and provides the user U with various information about real estate stored in the real estate information storage unit of the information processing device 10 through data communication with the information processing device 10 . The user terminals 50 are various client terminals having communication functions, such as smart phones, tablet terminals, personal computers, and wearable terminals.

For example, when the user U accesses the real estate site provided by the information processing device 10 and searches for a property, the user U operates the user terminal 50 to specify the conditions of the desired property. At this time, a method of suggesting search conditions and the like that the user U is highly likely to be interested in from the real estate site side based on the user U's search history and the like is also conceivable. However, for example, the proposed search conditions and the like are not necessarily effective search conditions.

Therefore, in the information processing according to the embodiment, the degree of influence that the presence or absence of clicks for each type of search condition has on conversion is calculated, and a search condition corresponding to the degree of influence is proposed. Note that the search condition type is, for example, various character strings constituting the search condition, for example, a keyword used at the time of searching.

As shown in FIG. 1, first, the information processing apparatus 10 performs an influence calculation process for calculating the influence of the presence or absence of clicks on conversion for each search condition type (step S01).

For example, the information processing apparatus 10 analyzes the presence or absence of click logs stored in the click log storage unit using a random forest, thereby calculating the degree of influence of the presence or absence of clicks on conversion for each search condition type. A click log is data that associates the presence or absence of clicks for each search condition type with the presence or absence of conversions.

The conversion here means an inquiry about a property by the user U for individual real estate information displayed as a search condition. may be a contract.

Random forest analysis of click logs automatically generates a decision tree T from the viewpoint of whether or not a search condition type is selected, and for the terminal node of the decision tree T, "This search condition type is selected If selected/not selected, the degree of influence related to conversion was/did not lead to conversion is calculated as a score.

That is, in the random forest here, the correlation between the presence/absence of selection (presence/absence of clicks) and the presence/absence of conversion for each search condition type can be derived by machine learning as the degree of influence.

After that, when the information processing device 10 receives an access request to the real estate site from the user terminal 50 (step S02), the information processing device 10 performs processing for setting search conditions to be displayed on the user terminal 50 (step S03).

For example, the information processing apparatus 10 sets a search condition that combines the search condition types having a high degree of influence calculated in step S01. In the example shown in the same figure, "better than market rent", "low-rise property", and "highly rated property" are shown as search condition types, and "better than market rent", "low-rise property", and "user It shows that the degree of impact decreases in the order of "property with high evaluation".

For example, the information processing apparatus 10 sets a new search condition such as "a low-rise property at a better price than the market rent" by combining two search condition types, "a better price than the market price" and "a low-rise property." Note that one search condition type having a high degree of influence may be set as a search condition, or three or more search condition types may be combined to set a new search condition.

Subsequently, the information processing apparatus 10 proposes the set search conditions to the user U by transmitting the set search conditions to the user terminal 50 (step S04). As a result, the user terminal 50 displays search conditions set in consideration of the degree of impact on conversion. Here, as shown in FIG. 1, for example, a search condition proposed by the information processing apparatus 10 is displayed on the user terminal 50 as one of the search categories. That is, a search condition having a large influence on conversion is newly generated as one of the search categories, and the user U can select the search condition generated as one of the search categories.

As described above, the information processing apparatus 10 according to the embodiment can propose search conditions for which a high conversion is expected by proposing search conditions of a search condition type according to the degree of influence. Thereby, according to the information processing apparatus 10 according to the embodiment, it is possible to optimize the proposed search condition.

[2. Configuration example of information processing device]
Next, a configuration example of the information processing apparatus 10 will be described with reference to FIG. FIG. 2 is a block diagram showing a configuration example of the information processing device 10 according to the embodiment. As shown in FIG. 2, the information processing apparatus 10 has a communication section 110, a storage section 120, and a control section .

The communication unit 110 is realized by, for example, a NIC (Network Interface Card) or the like. The communication unit 110 uses networks such as various wireless communication networks such as 4G (Generation), 5G, LTE (Long Term Evolution), Wifi (registered trademark) or wireless LAN (Local Area Network), or various wired communication networks. Information is sent and received to and from an external device via the network.

The storage unit 120 is realized by, for example, a semiconductor memory device such as a RAM or flash memory, or a storage device such as a hard disk or an optical disk. The storage unit 120 also has a real estate information storage unit 121 , a click log storage unit 122 , a user information storage unit 123 , a user attribute storage unit 124 and an influence storage unit 125 .

The real estate information storage unit 121 stores various information related to real estate provided by the information processing device 10 . FIG. 3 is a diagram showing an example of information stored in the real estate information storage unit 121 according to the embodiment.

As shown in FIG. 3, for example, the real estate information storage unit 121 stores items of information such as "property ID", "property data", and "search query" in association with each other. "Property ID" is an identifier for identifying real estate, and "property data" is data related to the content of real estate (property name, address, age of building, area, distance to station, etc.) and created by real estate agents. This is data related to the surrounding environment (for example, a quiet residential area, etc.). Real estate data is, for example, data posted by a real estate agent, but may be data edited by information processing apparatus 10 based on the posted data.

For example, the information processing apparatus 10 adds information such as "a popular pub nearby" based on the address of the real estate, and information such as "excellent view" based on the altitude and surrounding buildings to the real estate data. good too.

A “search query” is a search query associated with corresponding real estate data. The search query may be extracted from the real estate data, or may be arbitrarily set on the information processing apparatus 10 side. That is, the information processing apparatus 10 may generate a search query from a different angle than the content posted by the real estate agent. For example, in this case, the information processing device 10 may generate a keyword extracted from word-of-mouth information posted by the user U as a search query.

Returning to the description of FIG. 2, the click log storage unit 122 will be described. The click log storage unit 122 stores information related to a log that is a search history by the user U. FIG. FIG. 4 is a diagram showing an example of information stored in the click log storage unit 122 according to the embodiment.

In the example shown in FIG. 4, the click log storage unit 122 stores items of information such as "user ID", "query 1", "query 2", and "conversion" in association with each other. “User ID” is an identifier for identifying the user U who uses the web service provided by the information processing device 10 .

“Query 1” and “query 2” indicate search queries, which are “1” when selected by the user U and “0” when not selected by the user U. FIG. "Conversion" indicates the presence/absence of conversion, with "1" if conversion occurs and "0" if no conversion occurs. As described above, the conversion is, for example, an inquiry about a property by the user U, but it may also be storage of conditions when searching for a property, registration of a property as a favorite, application for viewing, a lease agreement, or a sales contract. In this case, for example, the click information storage unit 122 may store the presence or absence of conversion for each type of conversion.

Returning to the description of FIG. 2, the user information storage unit 123 will be described. The user information storage unit 123 is a storage unit that stores user information regarding the user U. FIG. FIG. 5 is a diagram showing an example of information stored in the user information storage unit 123 according to the embodiment. In the example shown in FIG. 5, the user information storage unit 123 stores items such as "user ID", "annual income", "place of work", "site browsing history", and "card usage history" in association with each other. .

"User ID" indicates an identifier for identifying user U, and "annual income" indicates user U's annual income. In addition, the “place of work” indicates the place of work of the user U. FIG. For example, the information related to "annual income" and "work location" may be information registered by the user U in the information processing apparatus 10 via the user terminal 50. You may decide to divert it.

“Site browsing history” indicates the browsing history of the user U on the Internet. Note that the site browsing history includes, for example, the browsing history related to the real estate site provided by the information processing apparatus 10, the browsing history in the cooperating service (for example, portal site, etc.), and the information related to the search query in the cooperating service. You can also try to Also, the site browsing history may include, for example, the purchase history at the shopping site and the viewing history of news and videos.

"Card usage history" indicates the usage history of user U's credit card. For example, information related to "card usage history" is information acquired from a cooperating credit card company. included. In addition to the card usage history, the user information storage unit 123 may store the payment history of other electronic payments (for example, QR code (registered trademark) payment, etc.).

Returning to the description of FIG. 2, the user attribute storage unit 124 will be described. The user attribute storage unit 124 is a storage unit that stores information about user U attributes. A user attribute here is the information regarding the user's U characteristic.

FIG. 6 is a diagram showing an example of information stored in the user attribute storage unit 124 according to the embodiment. In the example shown in FIG. 6, the user attribute storage unit 124 stores items such as "user ID", "annual income", "family structure", "lifestyle", and "general" in association with each other.

"User ID" indicates an identifier for identifying the user U. "Annual Income" indicates an annual income attribute of the user U, and attributes are classified according to the annual income. Note that the annual income attribute is, for example, in units of 1,000,000 yen. "Family structure" indicates the family structure attributes of the user U, and the attributes are classified according to, for example, the age and gender of all members of the family.

“Lifestyle” indicates attributes related to the user U's lifestyle, and the attributes are classified according to information related to the user U's age, sex, hobbies, preferences, and the like. Attributes related to lifestyle are classified according to various information related to, for example, "site browsing history" and "card usage history".

"Total" indicates a user attribute obtained by comprehensively viewing each attribute of the corresponding user U, for example. By quantifying the attributes of each item, such as annual income, family composition, and lifestyle, the attributes related to the overall are classified.

Returning to the description of FIG. 2, the impact storage unit 125 will be described. The impact storage unit 125 stores the impact on conversion for each search type. FIG. 7 is a diagram showing an example of information stored in the impact storage unit 125 according to the embodiment.

In the example shown in FIG. 7, the influence storage unit 125 stores items of information such as "search axis 1", "search axis 2", "influence", "positive", and "negative" in association with each other. "Search axis 1" and "search axis 2" indicate search queries. For example, "influence" indicates the influence of search axis 1 and search axis 2 on conversion. Note that the example of FIG. 7 shows a case where the degree of influence is displayed as a score SC1 or a score SC2.

"Positive" and "Negative" indicate the breakdown of impact. For example, positive indicates the degree to which it had a positive effect on conversions, i.e. how much it promoted conversions, and negative indicates the degree to which it had a negative effect on conversions, i.e. how much it inhibited clicks. It shows an index such as

Returning to the description of FIG. 2, the control unit 130 will be described. The control unit 130 is, for example, a controller, and a CPU (Central Processing Unit), MPU (Micro Processing Unit), or the like allows various programs stored in a storage device inside the information processing apparatus 10 to use the RAM as a work area. It is realized by executing as Also, the control unit 130 is a controller, and is implemented by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

As shown in FIG. 2, the control unit 130 includes a real estate information acquisition unit 131, a user information acquisition unit 132, a behavior information acquisition unit 133, a user attribute classification unit 134, a calculation unit 135, a proposal unit 136, and a search unit 137 .

The real estate information acquisition unit 131 acquires real estate information to be published on a web page. For example, the real estate information acquisition unit 131 acquires real estate information submitted by a real estate agent and stores it in the real estate information storage unit 121 .

The user information acquisition unit 132 acquires various types of user information regarding the user U. FIG. For example, the user information acquisition unit 132 acquires user information from various cooperating services. For example, the user information acquisition unit 132 acquires information on family composition, annual income, work location (location of the company where you work), and credit card usage history from the credit card company. In addition, for example, the user information acquisition unit 132 acquires information such as web pages viewed by the user U through the portal site and search queries input on the portal site from the portal site provider.

Note that the user information acquisition unit 132 may acquire, for example, a history of position information from the user terminal 50 as user information. Also, the user information acquisition unit 132 writes the acquired various user information to the user information storage unit 123 .

The behavior information acquisition unit 133 acquires behavior information about the behavior of the user U performed on the search result for each search condition type. For example, the behavior information acquisition unit 133 acquires, as behavior information, information on the search condition type selected by each user U on the real estate site provided by the information processing apparatus 10 and information on the presence or absence of conversion for the search result of the search type. The action information acquisition unit 133 stores the acquired action information in the click log storage unit 122 each time the action information is acquired.

The user attribute classification unit 134 classifies the attributes of the user U based on the user information. For example, the user attribute classification unit 134 classifies the user U attributes based on the user information stored in the user information storage unit 123 .

For example, the user attribute classification unit 134 classifies the annual income attribute of the user U from the user information regarding annual income, and classifies the family composition attribute of the user U from the user attribute regarding family composition. Also, the user attribute classification unit 134 classifies the lifestyle attributes of the user U based on user information related to site browsing history and the like. In addition, the user attribute classification unit 134 classifies the general attributes of the user U using an algorithm for comprehensively analyzing these attributes.

The calculation unit 135 uses the behavior information acquired by the behavior information acquisition unit 133 to calculate the degree of influence of the search condition type on conversion. For example, the calculation unit 135 calculates the degree of impact by performing machine learning using random forest with click logs as input data.

At this time, in the random forest, a decision tree T with each search query as a node is automatically generated, and for the terminal node, "If this search query was selected/not selected, this conversion A score is given for "connected/not connected to". A score that leads to a conversion corresponds to a positive impact, and a score that does not lead to a conversion corresponds to a negative impact.

In the random forest, the degree of impact is calculated by using as input data information about search queries selected by the user U when conversion occurred and search queries not selected. For example, in the decision tree T, the search query corresponding to each node is ``selected'' or ``not selected'', and a score regarding positive or negative influence is output to the terminal node. .

At this time, when the user U selects a plurality of search queries and results in a conversion, the calculation unit 135 executes random forest using the plurality of search queries as one input data. That is, in this case, the multiple search queries form one node of the decision tree T in the random forest.

For example, when search query A and search query B are selected by user U and lead to conversion, search query A and search query B are treated as one search condition type.

For example, in this case, the respective search condition types of search query A and search query B are regarded as not being individually selected, and the search condition with search query A and search query B as one search condition type As a result of selecting the type, it is considered that conversion has occurred, and random forest is executed.

As a result, when a plurality of search condition types are selected in the process leading to conversion and conversion is reached, the calculation unit 135 appropriately calculates the influence of each of the individual search condition types and the plurality of search condition types. can be calculated to

Further, the calculation unit 135 may calculate the degree of influence for each attribute of the user U, for example. For example, the calculation unit 135 refers to the user attribute storage unit 124, extracts a click log corresponding to a predetermined user attribute from the click log storage unit 122, and executes a random forest using the extracted click log. Calculate the impact of each search condition type on conversion for each user attribute.

That is, the calculating unit 135 can derive the influence according to the user attribute by calculating the influence for each user attribute. This makes it possible to calculate the degree of influence more precisely than when calculating the degree of influence using the click logs of all users U.

The proposing unit 136 proposes to the user U search conditions of the search condition type selected according to the degree of influence calculated by the calculating unit 135 . The proposing unit 136 refers to the influence storage unit 125, selects search conditions that combine search queries with high influence scores, and provides the user terminal 50 of the user U with the search conditions.

At this time, the proposing unit 136 may propose a search condition including a search query different from the search conditions that the user U can input. In this case, for example, the search conditions that the user U can input include items related to living space such as property and location, such as rent, floor plan (including size), area, building age, and distance from a station.

On the other hand, the proposal unit 136 proposes an additional condition as a search condition from a different perspective from the search conditions that the user U can input. For example, as an example of additional conditions here, conditions related to the surrounding environment correspond. and so on.

For example, as described above, by suggesting additional search conditions, the user U can recall more specific conditions for property selection.

For example, the proposing unit 136 refers to the influence storage unit 125, generates a search condition by combining a plurality of search queries whose influence is equal to or greater than a predetermined value, and provides the user U with the search condition. At this time, the proposing unit 136 may, for example, propose a search condition that combines search queries having a positive influence.

That is, by proposing a search condition that combines search queries that mutually positively affect conversion, a synergistic effect between the search queries can be expected to have a positive effect.

Also, the proposing unit 136 may propose search conditions that combine search queries having different concepts, for example. The different concepts here are, for example, concepts classified by a predetermined classification method into rent, building age, floor plan, area, etc., and the proposal unit 136 proposes search conditions that combine search queries with different concepts. do. In other words, search conditions are proposed by avoiding combinations of search queries with the same concept.

For example, consider a search query generated from word-of-mouth for "highly rated by users" and "highly rated shared facilities and amenities." In this case, "high user evaluation" is, for example, a degree of satisfaction as a comprehensive evaluation of the property, and is a concept that includes "high evaluation of shared facilities and equipment."

For this reason, even if both search queries are combined, the contents of appeals overlap, which is not preferable. For this reason, the proposal unit 136 can generate new search conditions by combining search queries of different concepts whose appeal contents do not overlap. , it is possible to propose new search conditions that have not existed in the past.

Further, for example, when the calculating unit 135 calculates the degree of influence for each user attribute, the proposing unit 136 proposes a search condition based on the degree of influence corresponding to the user U's user attribute.

This makes it possible to propose more effective search conditions to the user U from the viewpoint of conversion. Also, the proposing unit 136 may propose, for example, search conditions different from the search conditions currently displayed on the user terminal 50 . That is, the proposal unit 136 may propose another search condition when the screen displayed on the user terminal 50 changes. As a result, a new search condition is displayed on the user terminal 50 for each screen transition, so various search conditions can be proposed to the user. Note that the proposal unit 136 may provide a history of proposed search conditions in response to a request from the user terminal 50, for example.

Moreover, the proposal unit 136 may propose the search condition when the number of pieces of real estate information matching the search condition to be proposed is greater than a predetermined number. That is, if the number of search results obtained by a proposed search condition is less than a predetermined number, the search condition may not be proposed. In this case, the proposing unit 136 may perform search processing by the searching unit 137 before proposing a search condition, and then determine whether or not to propose the search condition according to the search result.

In addition, the proposal unit 136, for example, after analyzing the tendency of the property information viewed by the user U, for example, proposes a search condition that includes properties that the user U is highly likely to convert. good. For example, in this case, the proposal unit 136 may propose a search condition that includes a search query associated with the corresponding property.

In addition, the proposal unit 136 may propose, for example, a more personalized search condition to the user U. For example, a more personalized search condition may be added to a search condition that combines multiple search queries.

For example, the proposal unit 136 generates individual search conditions personalized from the user U's search history, user attributes, and the like. As a more specific example, when proposing a search condition such as "highly rated tower apartment that suits you", the above search conditions are "perfect for you", "highly rated", and "tower apartment". decomposed into one search query.

Of these, the proposal unit 136 sets, for example, a property that corresponds to "perfect for you" according to the search history of the user U, user attributes, and the like. For example, the proposal unit 136 analyzes the conditions desired by the user U when selecting a property from the search history of the user U, sets the property that meets the conditions, or sets the probability of conversion in the user attribute to which the user U belongs. Set properties that meet high conditions.

Thereby, the user U can specify his/her desired search conditions only by selecting the proposed search conditions, and can easily designate search conditions that meet his/her desires.

The search unit 137 searches for real estate information according to search conditions set by the user U, and transmits the search results to the user terminal 50 . For example, the search unit 137 searches the real estate information storage unit 121 based on search conditions received from the user terminal 50, and outputs information about real estate that matches the search conditions as search results.

[3. Processing flow]
Next, an example of a processing procedure executed by the information processing apparatus 10 according to the embodiment will be described with reference to FIGS. 8 and 9. FIG. FIG. 8 is a flowchart illustrating an example of influence degree calculation processing according to the embodiment. FIG. 9 is a flowchart illustrating an example of search condition proposal processing according to the embodiment.

As shown in FIG. 8, first, the information processing device 10 acquires a click log (step S101). Subsequently, for example, when the number of newly acquired click logs reaches a predetermined number, the information processing apparatus 10 uses the click logs to calculate the degree of impact for each search query (step S102), and ends the process.

Next, a processing procedure for search condition processing will be described with reference to FIG. As shown in FIG. 9, first, the information processing apparatus 10 acquires a display request regarding a search condition proposal screen from the user terminal 50 (step S201).

Subsequently, the information processing apparatus 10 selects a search query to be proposed as a search condition according to the degree of influence of each search query (step S202). Subsequently, the information processing apparatus 10 proposes search conditions including the selected search query (step S203), and ends the process.

[4. Modification]
In the above-described embodiment, the real estate site provided by the information processing apparatus 10 has described a case where search conditions are proposed in consideration of the degree of influence of each search query, but the present invention is not limited to this. The present invention may be applied to various search services including net shopping.

Further, in the above-described embodiment, the case where the presence or absence of an inquiry for an individual property is regarded as a conversion has been described, but the conversion is not limited to this. For example, various actions taken by the user U with respect to the search results, such as a contract for a property (purchase of goods) and a request for materials, may be regarded as conversions.

In addition, the user's actions on the search results are not limited to the actions within the site, but may include, for example, purchasing/did not purchase the products displayed as the search results at the actual store. may In this case, the information processing apparatus 10 may acquire action information from, for example, a card statement or a QR code (registered trademark) usage history, and then calculate the degree of impact on the action.

Other actions taken by the user regarding the search results, such as whether or not they left the site after the search results were displayed, or whether or not they accessed the site for a specified period of time after the search results were displayed, etc. It may be set arbitrarily according to the calculation target of the degree.

[5. effect〕
The information processing apparatus 10 according to the above-described embodiment includes the action information acquisition unit 133 that acquires action information related to the user's action performed on the search result for each search condition type, and the action information acquired by the action information acquisition unit 133 A calculation unit 135 that calculates the degree of influence of the selection of a search condition type on behavior using behavior information, and a search condition of the search condition type selected according to the degree of influence calculated by the calculation unit 135 is sent to the user. and a proposing unit 136 for proposing. Therefore, according to the information processing apparatus 10 according to the embodiment, it is possible to optimize the proposed search conditions.

Also, the behavior information acquisition unit 133 according to the above-described embodiment acquires conversion information related to conversion performed on search results as click information, and the calculation unit 135 calculates the degree of impact using the conversion information. Therefore, according to the information processing apparatus 10 according to the embodiment, since it is possible to calculate the degree of impact on conversion for each search condition type, for example, it is possible to propose a search condition with a high expected value of conversion.

Further, the calculation unit 135 according to the above-described embodiment calculates the degree of impact by performing machine learning using random forest. Therefore, according to the information processing apparatus 10 according to the embodiment, it is possible to accurately calculate the degree of influence.

In addition, the calculation unit 135 according to the above-described embodiment calculates the degree of impact using as input data information regarding the search condition type selected by the user and the search condition type not selected when conversion occurred. Therefore, according to the information processing apparatus 10 according to the embodiment, it is possible to appropriately calculate the degree of influence.

In addition, the calculation unit 135 according to the above-described embodiment, when the user selects a plurality of search condition types and leads to conversion, uses the plurality of search condition types as one search condition type and machine learning using random forest I do. Therefore, according to the information processing apparatus 10 according to the embodiment, it is possible to appropriately calculate the degree of influence even when selection of each search condition type overlaps in the process leading to conversion.

Further, the calculation unit 135 according to the above-described embodiment calculates a positive or negative influence degree for each search condition type, and the proposal unit 136 proposes a search condition combining the search condition types calculated as the positive influence degree. suggest. Therefore, according to the information processing apparatus 10 according to the embodiment, it is possible to propose a search condition that can be expected to have a positive effect due to a synergistic effect between search condition types having a positive influence.

Further, the proposal unit 136 according to the above-described embodiment proposes search conditions including search condition types different from search conditions that can be input by the user. Therefore, according to the information processing apparatus 10 according to the embodiment, it is possible to propose a search condition from a different perspective than the conditions that the user can input. can be proposed.

Also, the proposal unit 136 according to the above-described embodiment proposes a search condition that combines search condition types having different concepts. Therefore, according to the information processing apparatus 10 according to the embodiment, since it is possible to propose search conditions from various angles, for example, it is possible to propose search conditions from new angles that have not existed in the past.

Further, the calculation unit 135 according to the above-described embodiment calculates the degree of influence for each user attribute classified according to user information about the user. Therefore, according to the information processing apparatus 10 according to the embodiment, it is possible to calculate an appropriate degree of influence suitable for user attributes.

Also, the proposal unit 136 according to the above-described embodiment proposes search conditions according to the degree of influence calculated for each user attribute. Therefore, according to the information processing apparatus 10 according to the embodiment, it is possible to propose appropriate search conditions to the user.

Also, the behavior information acquisition unit 133 according to the above-described embodiment acquires behavior information of search results related to real estate content, and the proposal unit 136 proposes search conditions related to real estate content. Therefore, according to the information processing apparatus 10 according to the embodiment, it is possible to propose appropriate search conditions for real estate content.

Further, the information processing method according to the above-described embodiment is an information processing method executed by a computer, and includes a behavior information acquisition step of acquiring behavior information related to a user's behavior performed on search results for each search condition type. and a calculation step of calculating the degree of influence of the selection of the search condition type on the action using the behavior information obtained by the behavior information obtaining step, and a search selected according to the degree of influence calculated by the calculation step and a proposing step of proposing a search condition of the condition type to the user. Therefore, according to the information processing method according to the embodiment, it is possible to optimize the proposed search conditions.

Further, the information processing program according to the above-described embodiment includes: a behavior information acquisition procedure for acquiring behavior information related to a user's behavior performed on search results for each search condition type; Using action information, a calculation procedure for calculating the degree of influence of whether or not a search condition type is selected on behavior, and a search condition of the search condition type selected according to the degree of influence calculated by the calculation procedure are proposed to the user. The computer is caused to execute the proposed procedure to be performed. Therefore, according to the information processing program according to the embodiment, it is possible to optimize the proposed search conditions.

[6. Hardware configuration]
Also, the information processing apparatus 10 according to the above-described embodiments is implemented by a computer 1000 configured as shown in FIG. 10, for example. FIG. 10 is a hardware configuration diagram showing an example of a computer that implements the functions of the information processing apparatus 10 according to the embodiment. Computer 1000 has CPU 1100 , RAM 1200 , ROM 1300 , HDD 1400 , communication interface (I/F) 1500 , input/output interface (I/F) 1600 and media interface (I/F) 1700 .

The CPU 1100 operates based on programs stored in the ROM 1300 or HDD 1400 and controls each section. The ROM 1300 stores a boot program executed by the CPU 1100 when the computer 1000 is started up, a program depending on the hardware of the computer 1000, and the like.

The HDD 1400 stores programs executed by the CPU 1100, data used by the programs, and the like. Communication interface 1500 receives data from other devices via network (communication network) N, sends the data to CPU 1100, and transmits data generated by CPU 1100 to other devices via network N. FIG.

Through an input/output interface 1600, the CPU 1100 connects an output device such as a display and a printer, and an input device such as a keyboard and a mouse (in FIG. 10, the output device and the input device are collectively referred to as the "input/output device") ). CPU 1100 acquires data from an input device via input/output interface 1600 . CPU 1100 also outputs the generated data to an output device via input/output interface 1600 .

Media interface 1700 reads programs or data stored in recording medium 1800 and provides them to CPU 1100 via RAM 1200 . CPU 1100 loads such a program from recording medium 1800 onto RAM 1200 via media interface 1700, and executes the loaded program. The recording medium 1800 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or a PD (Phase change rewritable disc), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory. etc.

For example, when the computer 1000 functions as the information processing apparatus 10 according to the embodiment, the CPU 1100 of the computer 1000 implements the functions of the control unit 130 by executing programs loaded on the RAM 1200 . The CPU 1100 of the computer 1000 reads these programs from the recording medium 1800 and executes them, but as another example, these programs may be acquired via the network N from another device.

As described above, some of the embodiments of the present application have been described in detail based on the drawings. It is possible to carry out the invention in other forms with modifications.

[7. others〕
Further, among the processes described in the above embodiments and modifications, all or part of the processes described as being performed automatically can be performed manually, or described as being performed manually. All or part of the processing can also be performed automatically by known methods. In addition, information including processing procedures, specific names, various data and parameters shown in the above documents and drawings can be arbitrarily changed unless otherwise specified. For example, the various information shown in each drawing is not limited to the illustrated information.

Also, each component of each device illustrated is functionally conceptual, and does not necessarily need to be physically configured as illustrated. In other words, the specific form of distribution and integration of each device is not limited to the one shown in the figure, and all or part of them can be functionally or physically distributed and integrated in arbitrary units according to various loads and usage conditions. Can be integrated and configured.

Also, the above-described embodiments and modifications can be appropriately combined within a range that does not contradict the processing content.

Also, the above-mentioned "section, module, unit" can be read as "means" or "circuit". For example, the real estate information acquisition unit 131 can be read as real estate information acquisition means or a real estate information acquisition circuit.

10 information processing device 50 user terminal 110 communication unit 120 storage unit 121 real estate information storage unit 122 click log storage unit 123 user information storage unit 124 user attribute storage unit 125 influence storage unit 130 control unit 131 real estate information acquisition unit 132 user information acquisition Unit 133 Action information acquisition unit 134 User attribute classification unit 135 Calculation unit 136 Proposal unit 137 Search unit

Claims (13)

an action information acquisition unit that acquires action information related to user actions performed on search results for each search condition type;
a calculation unit that uses the behavior information acquired by the behavior information acquisition unit to calculate the degree of influence that the presence or absence of selection of the search condition type has on the behavior;
and a proposing unit that proposes to a user search conditions of the search condition type selected according to the degree of influence calculated by the calculating unit. The behavior information acquisition unit
Acquiring conversion information related to the conversion performed on the search result as the action information;
The calculation unit
The information processing apparatus according to claim 1, wherein the degree of influence is calculated using the conversion information. The calculation unit
The information processing apparatus according to claim 2, wherein the degree of influence is calculated by performing machine learning using random forest. The calculation unit
4. The information according to claim 3, wherein the degree of impact is calculated using information about the search condition type selected by the user and the search condition type not selected by the user when the conversion occurred. processing equipment. The calculation unit
When the user selects a plurality of the search condition types and leads to the conversion, machine learning is performed using the random forest with the plurality of search condition types as one search condition type. The information processing apparatus according to claim 3 or 4. The calculation unit
calculating the positive or negative impact for each of the search condition types;
The proposal department
The information processing apparatus according to any one of claims 1 to 5, wherein the search condition is proposed by combining the search condition types calculated as the positive influence. The proposal department
The information processing apparatus according to any one of claims 1 to 6, wherein the search condition including the search condition type different from search conditions that can be input by the user is proposed. The proposal department
The information processing apparatus according to any one of claims 1 to 7, wherein the search condition is proposed by combining the search condition types having different concepts. The calculation unit
The information processing apparatus according to any one of claims 1 to 8, wherein the degree of influence is calculated for each user attribute classified according to user information about the user. The proposal department
The information processing apparatus according to claim 9, wherein the search condition is proposed according to the degree of influence calculated for each user attribute. The behavior information acquisition unit
obtaining the action information performed on the search results for real estate content;
The proposal department
The information processing apparatus according to any one of claims 1 to 9, wherein the search condition regarding the real estate content is proposed. A computer-executed information processing method comprising:
an action information acquisition step of acquiring action information related to user actions performed on search results for each search condition type;
a calculation step of calculating, using the behavior information obtained by the behavior information obtaining step, the degree of influence of the presence or absence of selection of the search condition type on the behavior;
and a proposing step of proposing to a user the search condition of the search condition type selected according to the degree of influence calculated in the calculating step. an action information acquisition procedure for acquiring action information relating to user actions performed on search results for each search condition type;
a calculating step for calculating, using the action information acquired by the action information acquiring step, the degree of influence of the presence or absence of selection of the search condition type on the action;
and a proposal procedure for proposing to a user search conditions of the search condition type selected according to the degree of influence calculated by the calculation procedure.

Images