JP6373462B1 - Prediction device, prediction method, and prediction program - Google Patents

Abstract

A prediction device, a prediction method, and a prediction program for improving the efficiency of the entire physical distribution are provided.
The prediction apparatus includes an acquisition unit and a prediction unit. The acquisition unit acquires action information associated with a predetermined area, which is user action information. The predicting unit predicts a delivery amount of a product that is a target of the commercial transaction and generated in a predetermined area based on the behavior information acquired by the acquisition unit. For example, the acquisition unit acquires, as behavior information, information related to an order for a provider that provides a product, and the order is for a predetermined area. The predicting unit predicts the delivery amount generated in the predetermined area based on the information related to the order.
[Selection] Figure 1

Description

  The present invention relates to a prediction device, a prediction method, and a prediction program.

  Conventionally, various commercial transactions are performed via the Internet. For example, on the Internet, a platform is provided that mediates a commercial transaction between a business operator who operates a store and a user, and a commercial transaction between users.

  As a technique related to the commercial transaction as described above, a technique for improving efficiency related to shipping by managing a package that is supposed to be shipped is known.

Japanese Patent No. 4938682

  However, with the above-described conventional technology, it is difficult to improve the efficiency of the entire physical distribution. For example, the prior art only estimates packages shipped using customer data of companies, and does not estimate logistics that can occur in the world.

  The present application has been made in view of the above, and an object thereof is to provide a prediction device, a prediction method, and a prediction program capable of improving the efficiency of the entire physical distribution.

  The prediction device according to the present application is user behavior information, an acquisition unit that acquires behavior information associated with a predetermined area, and a product that is a target of a commercial transaction based on the behavior information acquired by the acquisition unit And a predicting unit for predicting the delivery amount generated in the predetermined area.

  According to one aspect of the embodiment, there is an effect that the efficiency of the entire physical distribution can be improved.

FIG. 1 is a diagram illustrating an example of a prediction process according to the embodiment. FIG. 2 is a diagram illustrating a configuration example of the prediction system according to the embodiment. FIG. 3 is a diagram illustrating a configuration example of the prediction device according to the embodiment. FIG. 4 is a diagram illustrating an example of a delivery history storage unit according to the embodiment. FIG. 5 is a diagram illustrating an example of a model storage unit according to the embodiment. FIG. 6 is a diagram illustrating an example of the behavior information storage unit according to the embodiment. FIG. 7 is a diagram illustrating an example of the provision information storage unit according to the embodiment. FIG. 8 is a flowchart (1) illustrating a processing procedure performed by the prediction apparatus according to the embodiment. FIG. 9 is a flowchart (2) illustrating a processing procedure performed by the prediction apparatus according to the embodiment. FIG. 10 is a hardware configuration diagram illustrating an example of a computer that realizes the function of the prediction device.

  Hereinafter, modes for carrying out a prediction device, a prediction method, and a prediction program according to the present application (hereinafter referred to as “embodiments”) will be described in detail with reference to the drawings. In addition, the prediction apparatus, the prediction method, and the prediction program which concern on this application are not limited by this embodiment. In the following embodiments, the same portions are denoted by the same reference numerals, and redundant description is omitted.

[1. Example of prediction process)
First, an example of a prediction process according to the embodiment will be described with reference to FIG. FIG. 1 is a diagram illustrating an example of a prediction process according to the embodiment. In FIG. 1, the prediction process performed by the prediction apparatus 100 which concerns on embodiment is demonstrated. Specifically, in FIG. 1, based on user behavior information associated with a predetermined area by the prediction device 100, the amount of merchandise to be subjected to commercial transactions, which is generated in the predetermined area. An example in which the process of predicting is performed will be described.

  In the example illustrated in FIG. 1, the prediction device 100, the user terminal 10, the web server 30, and the delivery company device 50 are connected to be communicable via a communication network (for example, the Internet) not shown.

  The prediction device 100 shown in FIG. 1 predicts a delivery amount of a product that is a target of a commercial transaction based on user behavior information associated with a predetermined area on a map, and the amount of delivery that occurs in the predetermined area. It is the server device which performs the processing which performs. The prediction device 100 predicts, for example, a delivery amount of a commodity traded on a predetermined commercial transaction site (a platform such as a shopping mall (electronic shopping mall) on a network) provided by the web server 30 as an object of a commercial transaction. . In the embodiment, as an example of a commercial transaction, a product delivered from each store that opens in a shopping mall is taken as an example. However, the product is not limited to this example. For example, the product is listed on an individual transaction or an auction site. It may be an exhibit or the like.

  A user terminal 10 illustrated in FIG. 1 is an information processing apparatus used by a user. Specifically, the user terminal 10 allows a user to browse a web page, order a product on a website (for example, a shopping mall on a network), or post a product information evaluation (so-called review). Used for. Further, the user terminal 10 executes a process of inputting or transmitting a search query (hereinafter simply referred to as “query”) and a process of acquiring a search result in order to use the search service. In the example of FIG. 1, the user terminal 10 is, for example, a smartphone or a tablet terminal. Hereinafter, the user may be read as the user terminal 10. For example, the description “the user transmits information” may actually indicate a situation “the user terminal 10 used by the user transmits information”.

  The web server 30 shown in FIG. 1 is a server device that provides various contents to a user when accessed from the user terminal 10. In the example illustrated in FIG. 1, the web server 30 provides a shopping mall site on a network, an auction site between individuals, and the like. In a shopping mall provided by the web server 30, each business operator opens a store that handles products of various categories. Further, the web server 30 may provide a search service for searching a website, or provide a predetermined SNS (Social Networking Service) site or the like. In FIG. 1, only one web server 30 is illustrated, but a plurality of web servers 30 may exist and each may provide a different service.

  In addition, the web server 30 acquires information (hereinafter referred to as “behavior information”) related to user behavior on the network via a service to be provided. The behavior information is stored in, for example, a storage unit provided in the web server 30 or a predetermined external storage device. Specifically, the behavior information is information indicating the content that the user has used a service (such as a shopping service or a search service in the example of FIG. 1) provided from the web server 30. For example, the behavior information includes a product order in a shopping mall, a search behavior for a search service, a review from a user for each store in the shopping mall, a posting to an SNS, and the like. The web server 30 is not limited to the above example, and may acquire any information as long as it is behavior information on the network.

  The delivery company device 50 is a server device used by the delivery company. For example, the delivery company device 50 receives a request for delivery of a product that is a target of a commercial transaction. Then, the delivery company apparatus 50 generates delivery information that is information in which the date and time of delivery and the delivery destination are registered. The delivery company performs delivery to the delivery destination based on the delivery information generated by the delivery company apparatus 50. Although illustration in FIG. 1 is omitted, a plurality of delivery companies may exist. In the following, the delivery company may be read as the delivery company device 50. For example, the description “the delivery company transmits information” may actually indicate the situation “the delivery company device 50 used by the delivery company transmits information”.

  In addition, the delivery company device 50 provides information related to physical distribution to the prediction device 100. The information related to physical distribution is, for example, information related to a delivery amount generated at a predetermined date and time in the past and delivery resources required for delivery by a delivery company. The information related to the delivery resource is, for example, information such as the number of personnel, the number of vehicles, and the time required for delivery when delivering a predetermined delivery amount of goods. Further, the delivery company apparatus 50 may provide the prediction apparatus 100 with information on a base for actually delivering a product to the user. For example, when a predetermined delivery amount occurs in a predetermined area (region), the delivery company apparatus 50 determines in which base the delivery products are concentrated, and the personnel and time required for handling the delivery products at such bases. Information such as how much is required is provided to the prediction device 100.

  Here, it is considered that there is a correlation between the behavior information on the network and the delivery amount. For example, if a certain keyword becomes a buzzword and the keyword is a product name of a certain product or a keyword reminiscent of a product, the sales of the product are expected to increase after a predetermined time. . In this case, the order of goods is frequently made to a store or the like that opens in a shopping mall on the network, and the goods are shipped from the store, so the delivery amount increases accordingly. Such a situation may occur in a relatively wide geographical area such as a country, or may occur in a more detailed area (in Japan, a unit of a prefecture or a municipality).

  Therefore, the prediction apparatus 100 according to the embodiment acquires behavior information that is user behavior information on the network and is associated with a predetermined region, and performs the prediction process based on the acquired information, thereby performing the predetermined processing. It is possible to predict the delivery amount generated in the area. For example, the prediction device 100 may calculate the delivery amount when a certain keyword becomes a buzzword based on the rate of increase in the number of searches for past buzzwords and the sales (or delivery amount) of products related to the buzzword. Generate a model to predict the growth rate. For example, the prediction device 100 continuously acquires user behavior information, inputs the acquired information into a model, and predicts a delivery amount in a predetermined area. Thereby, since the prediction device 100 can grasp in advance the delivery amount assumed to occur in the predetermined area, the entire logistics in the predetermined area such as providing information on necessary delivery resources to the delivery company in advance. Can be made more efficient. Hereinafter, the prediction process according to the embodiment will be described along the flow with reference to FIG. 1. In the example of FIG. 1, it is assumed that the prediction device 100 has already generated a model for predicting the delivery amount using past data.

  In the example illustrated in FIG. 1, a user located in the area 60 uses various services provided by the web server 30. That is, the user terminal 10 accesses the web server 30 according to the operation by the user. Although not shown in the example of FIG. 1, there are a plurality of users and user terminals 10, and a considerable number of accesses sufficient to obtain a sufficient amount of information for performing the prediction process according to the embodiment are provided. Suppose you are going.

  For example, the user uses a service such as a search service provided by the web server 30 (step S11). The user may use a shopping service such as a shopping mall provided by the web server 30, or may use a service such as SNS.

  The web server 30 acquires various information transmitted from the user terminal 10. For example, the web server 30 acquires information regarding a query input by a user in the search service. Moreover, the web server 30 acquires information indicating that the user has ordered a predetermined product in a store such as a shopping mall. Further, the web server 30 may acquire text data and the like posted from the user. The information posted by the user includes, for example, writing to SNS, posting of a message to a message application, transmitted mail, and the like.

  At this time, the web server 30 determines the area where the accessing user is located, and acquires information regarding the area where the user is located. For example, when the user performs user registration in the provided service, the web server 30 determines the user's location area based on the address information registered by the user. Moreover, the web server 30 may determine the area where the user is located based on the IP address of the accessed user terminal 10. In addition, when the user terminal 10 has a position information acquisition function such as GPS (Global Positioning System), the web server 30 determines the region where the user is located by acquiring the position information. Also good. In this case, it is assumed that the web server 30 has predetermined map data for collating the acquired position information with the map information. In addition, the area | region used for a prediction process may be arbitrarily defined by the prediction apparatus 100, for example, may be a publicly defined area | region (for example, a municipality etc.). Further, the action information associated with the predetermined area does not necessarily mean that the user is located in the area where the delivery is performed. For example, the action information associated with the predetermined area includes an action for ordering from a store located in the predetermined area. In this case, the user is not necessarily located in the predetermined area, and may place an order from a distance. As described above, the action information associated with the predetermined area does not necessarily mean the location of the user, but includes information related to various actions taken by the user, which is somehow related to the area where the delivery is performed.

  The prediction device 100 acquires the behavior information associated with the region on the map as described above from the web server 30 (step S12). For example, the prediction device 100 transmits a behavior information acquisition request to the web server 30 and acquires behavior information associated with an area on the map from the web server 30 in response to the acquisition request.

  And the prediction apparatus 100 estimates the delivery amount in the area | region (area | region 60 in the example of FIG. 1) made into a process target based on the acquired action information (step S13). Although details will be described later, for example, in the region 60, the prediction device 100 determines the number of orders for a predetermined product, the number of searches for a predetermined keyword, the rate of increase in the number of searches, the rate of posting of a predetermined keyword and the rate of increase in the number of posts. Based on the above, the delivery amount after a predetermined time (for example, 24 hours later) in the region 60 is predicted.

  Subsequently, the prediction device 100 provides the predicted delivery amount to the delivery company device 50 (step S14). At this time, the prediction device 100 may also provide information related to the delivery resource corresponding to the predicted delivery amount. The delivery company device 50 prepares a delivery resource corresponding to the delivery amount based on the information provided from the prediction device 100 (step S15). Specifically, the delivery company apparatus 50 increases the number of personnel at a predetermined base (for example, a distribution center or a warehouse) in the area 60 or increases the number of vehicles operating in the area 60.

  As described above, the prediction device 100 according to the embodiment acquires behavior information associated with a predetermined area, which is user behavior information, and based on the acquired behavior information, delivery of a product that is a target of a commercial transaction. The amount of delivery generated in a predetermined area is predicted.

  That is, the prediction device 100 according to the embodiment predicts a delivery amount that can occur in an area on a certain map based on a user's search behavior performed on the network, order behavior, and a tendency such as posting to SNS. To do. In other words, the prediction device 100 can convert trends in the world that cannot be known only by the delivery company and trends such as the user's interests into an index value called delivery amount and provide it to the delivery company. Thereby, the prediction apparatus 100 can aim at the efficiency improvement of the whole physical distribution in a predetermined area | region. Hereinafter, the configuration of the prediction device 100 that performs the above-described processing and the prediction system 1 including the prediction device 100 will be described in detail.

[2. (Prediction system configuration)
The configuration of the prediction system 1 including the prediction device 100 according to the embodiment will be described with reference to FIG. FIG. 2 is a diagram illustrating a configuration example of the prediction system 1 according to the embodiment. As illustrated in FIG. 2, the prediction system 1 according to the embodiment includes a user terminal 10, a web server 30, a delivery company device 50, and a prediction device 100. These various apparatuses are communicably connected via a network N (for example, the Internet) in a wired or wireless manner. Note that the number of devices included in the prediction system 1 is not limited to the number illustrated in FIG. 2. For example, the prediction system 1 may include a plurality of user terminals 10.

  The user terminal 10 is, for example, an information processing apparatus such as a smartphone, a desktop PC (Personal Computer), a notebook PC, a tablet terminal, a mobile phone, a PDA (Personal Digital Assistant), or a wearable device. It is. The user terminal 10 accesses the web server 30 according to an operation by the user, and acquires content (for example, a web page) from the web server 30. Then, the user terminal 10 displays the acquired web page on a display device (for example, a liquid crystal display).

  The web server 30 is a server device that provides various web pages when accessed from the user terminal 10. For example, the web server 30 provides a search service, a shopping service, and a web page related to SNS. The web page provided by the web server 30 is not limited to this example, and the web server 30 may be, for example, a news site, a weather forecast site, a shopping site, a finance (stock price) site, a route search site, a map providing site, a travel site. Various web pages related to restaurants introduction sites, web blogs, etc. may be provided.

  Moreover, the web server 30 may acquire the action information of the user on the network via each website. The behavior information on the network includes, for example, information on a query on a search site, order and purchase behavior on a shopping site, information on a review posted by a user, text data posted on an SNS site by a user, and the like. .

  The delivery company device 50 is a server device used by the delivery company. For example, when a product is ordered from the user to the store, the delivery company device 50 receives a request for shipping the product from the store. In addition, the delivery company device 50 provides the prediction device 100 with information related to the distribution of the product.

  As described above, the prediction device 100 acquires user behavior information that is user behavior information and is associated with a predetermined region, and based on the acquired behavior information, delivers a product that occurs in the predetermined region. It is a server device that predicts the amount. Note that the prediction device 100 may also serve as a configuration as the web server 30. That is, the prediction device 100 may provide the search service and the shopping mall site by the prediction device 100 itself, and may acquire behavior information related to the search service, behavior information related to the shopping mall site, and the like.

  Note that the prediction apparatus 100 may receive provision from the web server 30 regarding user behavior information, for example, behavior information using a predetermined program that crawls on the network and acquires information. May be acquired continuously.

[3. Configuration of prediction device]
Next, the configuration of the prediction apparatus 100 according to the embodiment will be described with reference to FIG. FIG. 3 is a diagram illustrating a configuration example of the prediction device 100 according to the embodiment. As illustrated in FIG. 3, the prediction device 100 includes a communication unit 110, a storage unit 120, and a control unit 130. The prediction device 100 includes an input unit (for example, a keyboard and a mouse) that receives various operations from an administrator who uses the prediction device 100, and a display unit (for example, a liquid crystal display) that displays various types of information. You may have.

(About the communication unit 110)
The communication unit 110 is realized by, for example, a NIC (Network Interface Card). The communication unit 110 is connected to a communication network by wire or wireless, and transmits and receives information to and from the user terminal 10 and the like via the communication network.

(About the storage unit 120)
The storage unit 120 is realized by, for example, a semiconductor memory device such as a RAM (Random Access Memory) or a flash memory, or a storage device such as a hard disk or an optical disk. The storage unit 120 according to the embodiment includes a delivery history storage unit 121, a model storage unit 122, a behavior information storage unit 123, and a provision information storage unit 124. Hereinafter, each storage unit will be described in order.

(About the delivery history storage unit 121)
The delivery history storage unit 121 stores history information indicating a relationship between user behavior information and information related to physical distribution (for example, delivery amount). For example, the prediction device 100 acquires various types of information from the web server 30 and the delivery company device 50 via the network N, and continuously stores the acquired information in the delivery history storage unit 121. When using the history information for processing, the prediction device 100 appropriately acquires history information from the delivery history storage unit 121. Here, FIG. 4 shows an example of the delivery history storage unit 121 according to the embodiment. As illustrated in FIG. 4, the delivery history storage unit 121 includes items such as “history information ID”, “total period”, “target area”, and “delivery history information”. The items of “delivery history information” include small items such as “product ID”, “related keyword information”, “search increase rate”, “keyword appearance increase rate”, “delivery increase rate”, and “delivery amount information”. Including.

  “History information ID” indicates identification information for identifying history information. In this specification, identification information as shown in FIG. 4 may be used as a reference symbol used in the description. For example, the history information identified by the history information ID “B01” may be referred to as “history information B01”.

  “Total period” indicates the length of the period in which the behavior information and the delivery amount are totaled. Although illustration in FIG. 4 is omitted, a specific date and time when the behavior information and the delivery amount are totaled may be stored in the item of the total period. The “target area” indicates an area in which delivery history information and action information are tabulated.

  “Delivery history information” indicates information indicating the relationship between past user behavior information and information related to physical distribution. In the example of FIG. 4, the delivery history information is generated for each product handled at the shopping mall site provided by the web server 30.

  “Product ID” indicates identification information for identifying a product. Note that the product shown in FIG. 4 may be information on a product in a certain service (for example, a shopping mall site), or a product that is the same product and is handled on another shopping site or the like. The total information may be used.

  “Related keyword information” indicates a keyword related to the product. A keyword related to a product refers to a keyword that associates a product with the keyword or arouses sales of the product with the keyword. For example, a keyword related to a product is associated with a word appearing in a search suggestion together with the product, a word appearing together with a product name in a message such as SNS or a predetermined document, or a product in a predetermined knowledge database. Words. The word associated with the product in the predetermined knowledge database corresponds to, for example, the name of the entertainer who advertised the product, the phrase used as a catch phrase of the product, the competitive product of the product, the manufacturer who developed the product, etc. . Further, the prediction device 100 may appropriately generate the related keyword information of the product by using the knowledge database, for example, associating the product “mask” with the keyword “cold”. Further, when the search increase rate of a certain keyword is extremely high, the prediction apparatus 100 may extract a product or the like in which the number of orders is extremely increased after the predetermined time, and perform a process of associating the extracted product and the keyword. . Further, the prediction device 100 may have a data table that holds information in which products and associated keyword information are associated with each other in the storage unit 120 separately from the delivery history storage unit 121. In FIG. 4, the information stored in the “related keyword information” is conceptually expressed as “K01”, but actually, a specific keyword or keyword group is stored in the item. Is done.

  “Search increase rate” indicates the rate at which the number of times that a keyword included in related keyword information is searched as a query has increased in the last 24 hours. “Keyword appearance increase rate” indicates a rate at which the number of appearances of the keyword included in the related keyword information in SNS or the like has increased in the last 24 hours.

  “Delivery increase rate” indicates a rate of increase in the delivery amount of a product in a certain period. Note that the delivery increase rate may be, for example, information aggregated in a period after the aggregation period. For example, the delivery increase rate may be an increase rate of delivery during 24 hours after 24 hours immediately after the total period, 48 hours later, one week later, or the like.

  “Delivery amount information” indicates a delivery amount in which the product is actually delivered. In FIG. 4, the information stored in the “delivery amount information” is conceptually expressed as “E01”, but actually, a specific numerical value of the delivery amount is stored in the item. . The delivery amount may be indicated in any unit. For example, the delivery amount may be indicated by the number of packages actually delivered to the user by the delivery company, or by the total number of orders placed from the store or the like to the delivery company and the volume of each product. Alternatively, it may be information indicating a delivery amount that the delivery company has in-house. Further, the delivery amount may be indicated by a numerical value or the like indicating a delivery resource prepared by the delivery company with respect to the number of requests for dispatch. That is, the delivery amount may be expressed in any format as long as the delivery company can grasp the delivery amount.

  That is, in FIG. 4, as an example of data stored in the delivery history storage unit 121, the history information B01 whose history information ID is “B01” has a total period of “24 hours” and the target area is “area 60”. ". The history information B01 indicates delivery history information related to the product C01 identified by the product ID “C01”, the related keyword information of the product C01 is “K01”, and the search increase rate during the counting period is “60”. % ”And the keyword appearance increase rate is“ 30% ”. Further, the delivery increase rate after such a search increase rate is observed is “80%”, and the delivery amount information is “E01”.

  Although illustration in FIG. 4 is omitted, the delivery history storage unit 121 may store various information related to the delivery history and the merchandise in addition to the information described above. For example, the delivery history storage unit 121 may store a page view (Page View) for the product in the counting period, the number of unique users who have accessed the product page, and the like.

(About the model storage unit 122)
The model storage unit 122 stores information related to a model for predicting the delivery amount. Here, FIG. 5 illustrates an example of the model storage unit 122 according to the embodiment. As illustrated in FIG. 5, the model storage unit 122 includes items such as “model ID”, “learning data”, and “target region”.

  “Model ID” indicates identification information for identifying a model. “Learning data” indicates learning data used to generate a model. In FIG. 5, the information stored in the “learning data” is conceptually represented as “F01”, but actually, the learning data items include specific information used for model generation. For example, the delivery history information stored in the delivery history storage unit 121 is stored. The “target area” indicates an area targeted by the model for predicting the delivery amount. Although FIG. 5 shows an example in which a model and a region are associated with each other, the model may be used to predict a delivery amount in a region different from a region targeted by learning data.

  That is, in FIG. 5, as an example of data stored in the model storage unit 122, the model M01 having the model ID “M01” has the learning data “F01” and the target area “area 60”. Is shown.

(About the behavior information storage unit 123)
The behavior information storage unit 123 stores user behavior information. Here, FIG. 6 illustrates an example of the behavior information storage unit 123 according to the embodiment. As illustrated in FIG. 6, the behavior information storage unit 123 includes items such as “total date”, “target region”, “order information”, and “behavior information”. The “behavior information” items include medium items such as “search” and “post”, “search information”, “keyword”, “change rate”, “post information”, “keyword”, and “change rate”. It has a small item.

  “Aggregation date” indicates a date on which the user's behavior information is aggregated. The “target area” indicates an area associated with the user behavior information. “Order information” indicates information that the web server 30 has received an order for a product from a user. In FIG. 6, the information stored in the “order information” is conceptually expressed as “G01”. However, in actuality, the item of the order information includes each store that ships products in the target area. The total number of products ordered by the user is stored.

  “Behavior information” indicates information regarding the user's behavior on the network. “Search” indicates information related to search among the behavior information. “Search information” indicates information related to a search performed by the user on the aggregation date and the target area. In FIG. 6, information stored in “search information” is conceptually expressed as “H01”, but in actuality, the search information items include the number of times the user has performed a search, the user Is stored, the number of times each query is input (that is, the number of searches), and the like.

  “Keyword” indicates a keyword input as a query in the search. “Change rate” indicates the change rate of the number of searches with respect to the previous aggregation. In FIG. 6, for the sake of explanation, only keywords with a relatively large change rate are shown, but actually, the behavior information storage unit 123 stores all the keywords searched by the user and the number of searches. Is done.

  “Post” indicates information related to posting in the action information. “Posting information” indicates information related to postings made by the user on the aggregation date and the target area. In FIG. 6, information stored in “post information” is conceptually written as “J01”, but in actuality, the post information item includes text data submitted by the user to the SNS. In addition, the text data of the review posted on the shopping mall site, the text data of the transmitted mail, etc., and the word information obtained by analyzing the text data and extracting the included keywords are stored. Note that the prediction device 100 uses an algorithm for scoring each word appearing in one text data, such as tf-idf (Term Frequency-Inverse Document Frequency), as a word to be extracted from the text data. Only high words may be extracted.

  “Keyword” indicates a keyword that appears in the posting. “Change rate” indicates the change rate of the number of appearances of the keyword with respect to the immediately preceding aggregation. In FIG. 6, for the sake of explanation, only keywords with a relatively large change rate are shown, but actually, the behavior information storage unit 123 stores all the keywords posted by the user and the number of appearances thereof. Is done.

  That is, in FIG. 6, as an example of data stored in the behavior information storage unit 123, the behavior information associated with the target region “region 60” on the aggregation date “August 3, 2017” G01 ", search information is" H01 ", and post information is" J01 ". For example, the keyword “AAA” indicates that the rate of change in the number of searches is “2000%”, and the keyword “EEE” indicates that the rate of change in the number of postings is “1300%”.

(Provided information storage unit 124)
The provided information storage unit 124 stores information provided from the prediction device 100 to the delivery company. Here, FIG. 7 illustrates an example of the provision information storage unit 124 according to the embodiment. As illustrated in FIG. 7, the provision information storage unit 124 includes items such as “provision information ID”, “prediction date”, “predicted delivery amount”, and “increase rate”.

  “Provision information ID” indicates identification information for identifying the provision information. The “prediction date” indicates the date on which the delivery amount is predicted. For example, the predicted date is a date corresponding to a predetermined time (for example, 24 hours later) from the date when the behavior information is aggregated. The “predicted delivery amount” indicates a delivery amount predicted on the prediction date. In FIG. 7, the information stored in the “predicted delivery amount” is conceptually expressed as “N01”, but actually, the predicted delivery amount item is a specific numerical value, For example, a numerical value expressed in the same format (unit) as the delivery amount information shown in FIG. 4 is stored. “Increase rate” indicates an increase rate of the delivery amount from the immediately preceding forecast date.

  That is, in FIG. 7, the provision information L01 identified by the provision information ID “L01” has the predicted delivery amount “N01” on the prediction date “August 4, 2017”, and the prediction date immediately before (for example, 2017 This indicates that the rate of increase in the delivery amount from August 3, 2011) includes information that is “250%”.

(About the control unit 130)
The control unit 130 is a controller, for example, various programs (an example of a prediction program) stored in a storage device inside the prediction device 100 by a CPU (Central Processing Unit), an MPU (Micro Processing Unit), or the like. This is realized by executing the RAM as a work area. The control unit 130 is a controller, and is realized by an integrated circuit such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA).

  As illustrated in FIG. 3, the control unit 130 according to the embodiment includes an acquisition unit 131, a generation unit 132, a prediction unit 133, and a provision unit 134, and has functions and functions of information processing described below. Realize or execute. Note that the internal configuration of the control unit 130 is not limited to the configuration illustrated in FIG. 3, and may be another configuration as long as the information processing described below is performed. Further, the connection relationship between the processing units included in the control unit 130 is not limited to the connection relationship illustrated in FIG. 3, and may be another connection relationship.

(About the acquisition unit 131)
The acquisition unit 131 acquires various types of information. For example, the acquisition unit 131 acquires user behavior information that is associated with a predetermined area. The action information associated with the predetermined area refers to action information associated with information such as in which area the action is performed. Further, it refers to action information associated with information such as which area the action is performed on (for example, a store located in which area). Specifically, the behavior information associated with a predetermined area on the map is the location information of the user terminal 10 that is the transmission source of the behavior information, the address or residence of the user who uses the user terminal 10 Alternatively, it is information in which the delivery destination and inventory management location of the product are associated with a predetermined area.

  For example, the acquisition unit 131 is an order for a provider (for example, a store at a shopping mall site, an exhibitor at an auction site, etc.) that provides a product as behavior information, and relates to an order having a predetermined area as a shipping destination. Get information. For example, the acquisition unit 131 acquires information such as the number of items ordered by the user in a predetermined area and the size and quantity of products related to the order. For example, information relating to orders in a store or the like can be an element for predicting the delivery amount. Note that the acquisition unit 131 may acquire information such as when the product related to the order is shipped. In this case, the prediction unit 133 to be described later estimates that the product will be delivered at the date and time of shipping, and predicts the delivery amount at the date and time.

  Moreover, the acquisition part 131 acquires the information regarding the search action relevant to goods, and the search action which the user located in a predetermined area | region performs as action information. For example, the acquisition unit 131 acquires the number of searches performed by a user located in a predetermined area.

  More specifically, the acquisition unit 131 acquires an increase rate of the number of searches using a keyword related to a product as a query as information on search behavior. For example, for an arbitrary product, the acquisition unit 131 searches for the product name as a query, searches for the manufacturer name that manufactures the product as a query, or a keyword associated with the product (for example, a product in the search suggestion (Number of keywords that co-occur with name) and the rate of increase.

  Moreover, the acquisition part 131 acquires the information regarding the posting from the user located in the predetermined area | region which is a posting relevant to goods as action information. For example, the acquisition unit 131 acquires the number of postings and the like related to products made by a user located in a predetermined area. The post related to the product refers to, for example, a post including a related word such as a product name or a manufacturer name in the message.

  Specifically, the acquisition unit 131 includes at least a message created by the user, a text posted by the user on the SNS, an email created by the user, and a review posted by the user on a platform that is a place for commerce as information related to the posting. Get one. In this case, as described above, the acquisition unit 131 may acquire only a word (keyword) having a high weight in a message or a review using a known algorithm such as tf-idf.

  In addition, the acquisition unit 131 may acquire the amount of delivery performed in the past and information related to the delivery resource as information related to physical distribution. For example, the acquisition unit 131 acquires information related to physical distribution from the delivery company device 50. In addition, the acquisition unit 131 may acquire the delivery amount of the delivery performed in the past, and may acquire user behavior information before a predetermined time (for example, 24 hours before) when the delivery has occurred. The generation unit 132 to be described later generates a model for predicting the delivery amount based on the relationship between the delivery amount and the behavior information as described above.

  In addition, the acquisition unit 131 may acquire information indicating that the user has an interest in the product directly, such as a transition in the number of reactions from the user to the product or a PV for the product. . Moreover, the acquisition part 131 may acquire the information etc. which the user purchased arbitrary goods not only on the information on a network but in a real store, for example. In this case, the acquisition unit 131 acquires information related to the product purchased by the user via a cash register system or the like in the actual store. The acquisition unit 131 stores the acquired information in the storage unit 120 as appropriate. For example, the acquisition unit 131 stores past delivery history and past behavior information in the delivery history storage unit 121. Further, the acquisition unit 131 stores the behavior information of the user to be aggregated in the behavior information storage unit 123.

  The method by which the acquisition unit 131 acquires information is not limited. For example, the acquisition unit 131 may acquire information provided from the web server 30 or the delivery company device 50, or may crawl on the Internet using a program such as a search robot used for a search engine or the like. Thus, various information may be acquired at any time, or the acquired information may be updated.

(About the generator 132)
The generation unit 132 generates a model for predicting the delivery amount generated in the predetermined area. For example, the generation unit 132 gives the behavior information to the delivery amount based on the behavior information acquired by the acquisition unit 131 and the change (increase rate) in the delivery amount after a predetermined time since the behavior information is aggregated. Generate a model that reflects the impact.

  For example, the generation unit 132 determines how much the delivery amount of the product increases or the actual delivery amount of the product according to an increase rate of a search for a keyword related to an arbitrary product or an increase rate of the keyword. Get the relationship of how long it was. For example, the generation unit 132 generates such a relationship for all products for which the acquisition unit 131 can acquire information from the web server 30. Then, the generation unit 132 uses statistical processing to calculate a relationship such as how much the search increase rate and the keyword appearance rate are increased and how much the delivery amount is increased as a calculation formula (model). Generate.

  For example, the generation unit 132 is a store that sells products to the user using the web server 30 and performs a model generation process as described above, and stores in a predetermined area (for example, delivery to a predetermined area). This is performed for all of the stores having a warehouse, and a model for predicting the delivery amount of the product in a predetermined area is generated.

  The generation unit 132 is not limited to the statistical method as described above, and may generate a model using a predetermined learning process or the like. In other words, the prediction device 100 can be any model as long as it can represent the relationship between the user behavior information associated with the predetermined area on the map and the delivery amount of the product generated in the predetermined area. May be generated. Further, the generation unit 132 may generate a model based on various user actions shown in FIGS. 4 and 5, for example, without limiting to actions such as search and posting.

  In addition, for example, the generation unit 132 may generate a model by learning with a regression equation using each of the behavior information as features, using a delivery amount related to delivery performed in a predetermined area in the past by a delivery company as a positive example. Good. For example, the generation unit 132 creates the following formula (1).

y = ω ₁ · x ₁ + ω ₂ · x ₂ + ω ₃ · x ₃ ... + ω _N · x _N (1) (N is an arbitrary number)

The above formula (1) is generated for each predetermined area, for example. In the above formula (1), “y” is substituted with the actual delivery amount acquired from the delivery company device 50. In the above formula (1), “x” is information (features) related to user behavior, and these correspond to explanatory variables in the regression equation. In the above formula (1), “ω” is a coefficient of “x” and indicates a predetermined weight value. Specifically, “ω ₁ ” is a weight value of “x ₁ ”, “ω ₂ ” is a weight value of “x ₂ ”, and “ω ₃ ” is a weight value of “x ₃ ”. It is. As described above, the above equation (1) is a variable including the explanatory variable “x” corresponding to the feature information extracted from the behavior information and the predetermined weight value “ω” (for example, “ω ₁ · x _1”). ). For example, in the above formula (1), “x ₁ ” is the “search increase rate” shown in FIG. 4, and “x ₂ ” is the “keyword appearance increase rate” shown in FIG.

The production | generation part 132 produces | generates the said Formula (1) for every predetermined area and every total period. Then, the generation unit 132 sets the generated expression as a machine learning sample. Then, the generation unit 132 derives a value corresponding to the predetermined weight value “ω” by performing calculation processing (regression learning) of a sample expression. That is, the generation unit 132 determines a predetermined weight value “ω” that satisfies the above formula (1). In other words, the generation unit 132 can determine the weight value “ω” indicating the influence of the predetermined explanatory variable on the objective variable “y”. For example, if the “search increase rate” greatly contributes to the event “delivery amount in a predetermined area” as compared with other variables, the weight value “ω ₁ ” corresponding to the “search increase rate”. As for the value of "", a large value is calculated as compared with other variables.

  In the above example, the search increase rate and the keyword appearance increase rate are shown as features, but these values may be converted to be used as explanatory variables by a predetermined process. For example, the generation unit 132 may normalize and use these numerical values by a known method. Further, in the generation process using the above formula (1), the left side is not set to “1” or “0” itself, but a predetermined error is assumed, and a value obtained by squaring the difference from the error is the minimum value. The optimal solution of “ω” may be calculated using a method such as a least square method that approximates to In addition, the generation unit 132 may make various design changes, for example, the actual delivery amount is not a positive example, but the number of products actually shipped in a store that ships from a predetermined area is a positive example. .

  In addition, although description was abbreviate | omitted in said example, the production | generation part 132 may produce | generate a model by making every action information into a feature. Moreover, the model generation method illustrated is an example, and the model generated by the generation unit 132 is not limited to the above example. In other words, the generation unit 132 is a model that indicates the relationship between the behavior information and the delivery amount, and outputs a delivery amount in a predetermined area after a predetermined time when newly aggregated behavior information is input. If so, a model different from the above example may be appropriately generated by combining known methods. For example, in the above example, an example in which a model is generated by regression analysis has been shown. However, the generation unit 132 may generate a model by other statistical processing, or a model using a neural network or the like. It may be generated. In the learning process (machine learning), various methods such as linear regression, logistic regression, support vector machine, decision tree, and hazard analysis may be used. Then, the generation unit 132 stores the generated model in the model storage unit 122.

(About the prediction unit 133)
Based on the behavior information acquired by the acquisition unit 131, the prediction unit 133 predicts a delivery amount of a product that is a target of a commercial transaction and that occurs in a predetermined area.

  For example, when the information about an order with a predetermined area as a shipping destination is acquired as the behavior information, the prediction unit 133 obtains an order for a provider that provides a product. Predict the amount of delivery that will occur. In this case, the prediction unit 133 may predict the delivery amount without using the model generated by the generation unit 132. That is, the predicting unit 133 may predict the delivery amount based on the number, capacity, volume, or the like of products ordered from a store or the like and scheduled to be shipped on the prediction date.

  Further, the prediction unit 133 predicts a delivery amount generated in the predetermined area based on information related to a search action executed by a user located in the predetermined area, which is a search action related to the product as the action information. For example, the prediction unit 133 predicts a delivery amount that occurs after a predetermined time since the search information is aggregated by inputting the aggregated search information to the model generated by the generation unit 132.

  Specifically, the prediction unit 133 predicts a delivery amount generated in a predetermined area based on a rate of increase in the number of searches using a keyword related to a product as a query as information on search behavior. Accordingly, the predicting unit 133 can predict the delivery amount by accurately reflecting how a keyword that has become a sudden trend such as buzzword affects delivery after a predetermined time.

  Further, the prediction unit 133 predicts a delivery amount generated in the predetermined area based on information related to the posting from the user located in the predetermined area as a post related to the product as the behavior information. For example, the prediction unit 133 predicts a delivery amount that occurs after a predetermined time since the posting information is aggregated by inputting the aggregated posting information to the model generated by the generation unit 132.

  Specifically, the prediction unit 133 includes at least the message created by the user, the text posted by the user on the SNS, the email created by the user, and the review posted by the user on the platform that is the place of commerce as information related to the posting. One of them is acquired, and a delivery amount generated in a predetermined area is predicted based on the number of appearances or the appearance rate of the keywords included in the acquired information and related to the product.

  Note that, when the prediction unit 133 performs prediction using the model generated by the generation unit 132, the prediction unit 133 calculates the delivery amount of the predetermined region based on the score output from the model. The predicting unit 133 may use the output score itself as a predicted delivery amount in a predetermined area, or perform processing such as normalizing the output score, and predict a delivery amount predicted after the processing. It is good. The prediction unit 133 stores information related to the predicted delivery amount in the provision information storage unit 124. Note that the prediction unit 133 may predict the delivery amount of the entire product in the predetermined area, or may predict the delivery amount related to each product.

(About the providing unit 134)
The providing unit 134 provides the delivery amount predicted by the prediction unit 133 to a delivery company that delivers products in a predetermined area.

  Further, the providing unit 134 provides information on the delivery resources that the delivery company has in the predetermined area based on the delivery amount predicted by the prediction unit 133. For example, the providing unit 134 should have a delivery resource with respect to a predicted delivery amount based on the relationship between the delivery resource information and the delivery resource information that the delivery company has provided in a predetermined area in the past. Such information is provided to the delivery company.

  The providing unit 134 may provide the predicted delivery amount not only to the delivery company but also to the service side such as a store. As a result, the service side can know in advance that orders will increase within a specified time.For example, prepare products in advance to make shipping easier, or prepare more part-time workers working in the warehouse. You can do it.

[4. Processing procedure)
Next, the procedure of the prediction process performed by the prediction apparatus 100 according to the embodiment will be described with reference to FIGS. FIG. 8 is a flowchart (1) illustrating a processing procedure performed by the prediction apparatus 100 according to the embodiment.

  As illustrated in FIG. 8, the prediction device 100 acquires past behavior information and a delivery amount from the delivery history storage unit 121 (step S <b> 101). And the prediction apparatus 100 produces | generates the model for estimating the delivery amount in a predetermined area | region based on the acquired information (step S102). And the prediction apparatus 100 stores the produced | generated model in the model memory | storage part 122 (step S103).

  Next, the procedure of the delivery amount prediction process performed by the prediction device 100 according to the embodiment will be described with reference to FIG. FIG. 9 is a flowchart (2) illustrating a processing procedure performed by the prediction apparatus 100 according to the embodiment.

  As illustrated in FIG. 9, the prediction device 100 determines whether or not action information in a region to be processed has been acquired (step S <b> 201). When the behavior information has not been acquired, the prediction device 100 stands by until it is acquired (for example, until a predetermined period of counting has elapsed) (step S201; No).

  On the other hand, when the behavior information is acquired (step S201; Yes), the prediction device 100 inputs the acquired behavior information into the model (step S202).

  And the prediction apparatus 100 estimates the delivery amount in the area | region of a process target based on the information output from the model (step S203). Furthermore, the prediction device 100 provides the predicted delivery amount to the delivery company (step S204).

[5. (Modification)
The prediction device 100 according to the above-described embodiment may be implemented in various different forms other than the above-described embodiment. Therefore, in the following, another embodiment of the prediction device 100 will be described.

[5-1. (Prediction processing using information other than behavior information)
The prediction device 100 may predict the delivery amount using not only behavior information but also various information. For example, when an arbitrary product is reported on a medium or the like that transmits to a predetermined area, it is assumed that the sales of the product increase and the delivery amount increases accordingly.

  Therefore, the prediction device 100 may generate a model that takes into account the influence of past reports on the delivery amount, and may predict the delivery amount using the generated model. That is, the prediction device 100 may acquire news report information related to a product in a predetermined area and predict a delivery amount generated in the predetermined area based on the acquired news report information. Thereby, the prediction device 100 can predict the delivery amount from a wider variety of viewpoints.

[5-2. (Use information in other areas)
For example, a trend of a keyword or a product may occur in another area away from a predetermined area, and the trend may be transmitted to the predetermined area. In this case, the prediction device 100 acquires, for example, an increase rate of the actual delivery amount in another area where the keyword or the product has been prevalent, and performs prediction processing in the predetermined area using the acquired information. May be used. Thereby, the prediction apparatus 100 can improve the precision of prediction.

[5-3. Data to be input to model)
In embodiment mentioned above, the prediction apparatus 100 showed the example which inputs action information into the produced | generated model and predicts a delivery amount. Here, the prediction device 100 may associate the acquisition timing of the information used to generate the model with the behavior information input to the model.

  Since the information acquired by the prediction device 100 includes information related to a commercial transaction site such as a shopping mall or the like, there is a possibility that the information is affected by time. For example, there may be a difference in the number of shipments for each store between the time when transactions are actively performed, such as the end of the year, and other times. Therefore, the prediction device 100 stores the time when information is acquired to generate a model, and when inputting behavior information to be processed, the time when the information used for generating the model is acquired. Enter information obtained at similar times. Thereby, the prediction apparatus 100 can perform the prediction process which considered the time influence of the score output.

  Moreover, the prediction apparatus 100 may acquire action information for each day of the week or for each weather (for example, weather or temperature). Moreover, the prediction apparatus 100 may acquire the past delivery amount for every day of the week or every weather. Then, the prediction device 100 may perform model generation processing and delivery amount prediction processing using day of the week information and weather information as features. That is, it is assumed that the delivery amount varies depending on factors such as a day of the week and factors caused by weather as well as a time factor. The prediction device 100 can improve the accuracy of prediction by acquiring behavior information and delivery amount information including such day information and weather information and using them for processing.

[5-4. Amount of information)
In the above embodiment, the prediction apparatus 100 has shown an example in which a model is generated based on various information that can be acquired on the network. Here, the prediction apparatus 100 may perform processing using only information acquired from general users who use various websites that exceeds a certain threshold.

  For example, reviews regarding products in a certain store, user posts, and the like may show a biased tendency if they are not based on a certain number of data. In this case, since the influence of biased data tends to affect the regression equation, the prediction device 100 may generate a model that cannot accurately calculate the delivery amount. For this reason, for example, the prediction apparatus 100 may use only data whose total number of reviews transmitted from the user for a certain product exceeds a certain number as data handled in the model prediction process. Thereby, the prediction device 100 can generate a model that outputs a highly reliable prediction result.

[5-5. Information variations)
In the said embodiment, the prediction apparatus 100 showed the example which produces | generates a model based on the user review etc. contributed to a shopping mall etc. Here, the variation of the information which the prediction apparatus 100 handles is demonstrated in detail.

  For example, the prediction device 100 may perform morphological analysis on the acquired word as information included in the user review and acquire information aggregated for each part of speech. Specifically, the prediction device 100 acquires a total result obtained by classifying noun categories into people, things, places, and the like. Further, the prediction device 100 may acquire a classification result such as whether each word has a positive attribute or a negative attribute, and a determination result such as an intention determination of the dialog when the message is in a dialog format.

  Moreover, the prediction apparatus 100 may acquire audio information transmitted from the user as an example of post information. That is, the prediction device 100 may acquire the text data corresponding to the posting information by recognizing a voice transmitted from a user who uses voice search or voice input.

  Moreover, the prediction apparatus 100 may extract only a specific part of speech regarding the acquired word. For example, the prediction device 100 may use only parts of speech that are assumed to be related to a product, such as nouns and adjectives, for processing. Thereby, the prediction apparatus 100 can reduce the processing load without reducing the accuracy of the model to be generated.

[6. Others]
In addition, among the processes described in the above embodiment, all or part of the processes described as being automatically performed can be performed manually, or the processes described as being performed manually can be performed. All or a part can be automatically performed by a known method. In addition, the processing procedures, specific names, and information including various data and parameters shown in the document and drawings can be arbitrarily changed unless otherwise specified.

  Further, each component of each illustrated apparatus is functionally conceptual, and does not necessarily need to be physically configured as illustrated. In other words, the specific form of distribution / integration of each device is not limited to that shown in the figure, and all or a part thereof may be functionally or physically distributed or arbitrarily distributed in arbitrary units according to various loads or usage conditions. Can be integrated and configured.

  For example, the information in the storage unit 120 illustrated in FIG. 3 may be stored in an external storage server or the like without being stored in the prediction device 100. In this case, the prediction device 100 obtains various stored information by accessing the storage server.

  In addition, for example, the above-described prediction device 100 obtains various types of information or accepts a store delivery amount calculation request, such as a front-end server side that mainly performs exchanges with external devices, prediction processing, and the like. May be distributed to the back-end server side that executes. In this case, for example, the front-end server includes at least an acquisition unit 131 and a provision unit 134. The back-end server includes at least a generation unit 132 and a prediction unit 133.

[7. Hardware configuration)
The prediction device 100, the user terminal 10, the web server 30, the delivery company device 50, and the like according to the above-described embodiment are realized by a computer 1000 having a configuration as illustrated in FIG. Hereinafter, the prediction apparatus 100 will be described as an example. FIG. 10 is a hardware configuration diagram illustrating an example of a computer 1000 that implements the functions of the prediction device 100. The computer 1000 includes a 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 a program stored in the ROM 1300 or the HDD 1400 and controls each unit. 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 a program executed by the CPU 1100, data used by the program, and the like. The communication interface 1500 receives data from other devices via the communication network 500 (corresponding to the network N shown in FIG. 2) and sends the data to the CPU 1100, and other data generated by the CPU 1100 via the communication network 500 To the device.

  The CPU 1100 controls an output device such as a display and a printer and an input device such as a keyboard and a mouse via the input / output interface 1600. The CPU 1100 acquires data from the input device via the input / output interface 1600. Further, the CPU 1100 outputs the data generated via the input / output interface 1600 to the output device.

  The media interface 1700 reads a program or data stored in the recording medium 1800 and provides it to the CPU 1100 via the RAM 1200. The CPU 1100 loads such a program from the recording medium 1800 onto the RAM 1200 via the 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 PD (Phase change rewritable disk), 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 prediction device 100, the CPU 1100 of the computer 1000 implements the function of the control unit 130 by executing a program loaded on the RAM 1200. In addition, each data in the storage unit 120 is stored in the HDD 1400. 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 from other devices via the communication network 500.

[8. effect〕
As described above, the prediction device 100 according to the embodiment includes the acquisition unit 131 and the prediction unit 133. The acquisition unit 131 acquires user behavior information, which is behavior information associated with a predetermined area. Based on the behavior information acquired by the acquisition unit 131, the prediction unit 133 predicts a delivery amount of a product that is a target of a commercial transaction and that occurs in a predetermined area.

  As described above, the prediction device 100 according to the embodiment uses the user's behavior information for processing, and based on trends such as the world trend and the user's interest that cannot be known only by the delivery company, Delivery volume can be predicted. Thereby, since the prediction apparatus 100 can predict the delivery amount of a predetermined area | region accurately, it can aim at the efficiency improvement of the whole physical distribution in a predetermined area | region.

  In addition, the acquisition unit 131 acquires, as the behavior information, information related to an order for a provider that provides a product and that is an order having a predetermined area as a shipping destination. The predicting unit 133 predicts a delivery amount generated in a predetermined area based on information related to the order.

  As described above, the prediction device 100 according to the embodiment can predict the delivery amount with high accuracy by predicting the delivery amount based on the order made from the user to the store or the like.

  Moreover, the acquisition part 131 acquires the information regarding the search action relevant to goods, and the search action which the user located in a predetermined area | region performs as action information. The prediction unit 133 predicts a delivery amount generated in a predetermined area based on information related to search behavior.

  As described above, the prediction device 100 according to the embodiment can predict the delivery amount with high accuracy by performing the prediction based on the search behavior assumed to reflect the user's consumption behavior.

  Moreover, the acquisition part 131 acquires the increase rate of the search frequency which makes the keyword relevant to goods a query as information regarding search action. The prediction unit 133 predicts a delivery amount generated in a predetermined area based on the increase rate of the number of searches.

  As described above, the prediction device 100 according to the embodiment is adapted to the movement of the world such as how much a product related to a buzzword is sold and how much a delivery amount is generated when a buzzword or the like occurs. Prediction processing can be performed.

  Moreover, the acquisition part 131 acquires the information regarding the posting from the user located in the predetermined area | region which is a posting relevant to goods as action information. The prediction unit 133 predicts a delivery amount that occurs in a predetermined area based on information related to posting.

  As described above, the prediction device 100 according to the embodiment can predict the delivery amount with high accuracy by performing the prediction based on the posting behavior assumed to reflect the interest of the user.

  The acquisition unit 131 also includes at least one of a message created by the user, a text posted by the user on the SNS, an email created by the user, and a review posted by the user on a platform that is a place for commerce as information related to the posting. get. The prediction unit 133 predicts a delivery amount generated in a predetermined region based on the number of appearances or the appearance rate of keywords that are included in the information acquired by the acquisition unit 131 and are related to the product.

  As described above, the prediction device 100 according to the embodiment can perform prediction processing according to the movement of the world by performing prediction based on keywords included in messages frequently issued from users.

  Further, the acquisition unit 131 acquires news report information related to products in a predetermined area. The prediction unit 133 predicts a delivery amount generated in a predetermined area based on the news report information.

  As described above, the prediction device 100 according to the embodiment can predict the delivery amount from various viewpoints by predicting the delivery amount generated when there is a report in addition to the user behavior information.

  The prediction apparatus 100 according to the embodiment further includes a providing unit 134 that provides the delivery amount predicted by the prediction unit 133 to a delivery company that delivers products in a predetermined area.

  As described above, the prediction apparatus 100 according to the embodiment can transmit the delivery amount to the delivery company in advance by providing the predicted information to the delivery company, so that the delivery process in the predetermined area is efficiently performed. be able to.

  Further, the providing unit 134 provides information on the delivery resources that the delivery company has in the predetermined area based on the delivery amount predicted by the prediction unit 133.

  As described above, the prediction apparatus 100 according to the embodiment provides information including the delivery resource, and thereby, for example, causes the delivery company to arrange a larger number of personnel at the base and further increase the efficiency of the delivery process. Can be achieved.

  The acquisition unit 131 acquires action information including day information or weather information. The prediction unit 133 predicts the delivery amount generated in the predetermined area based on the behavior information including day information or weather information acquired by the acquisition unit 131.

  As described above, the prediction device 100 according to the embodiment can predict the delivery amount with high accuracy by performing the prediction in consideration of the day of the week and the weather (such as weather).

  Further, the prediction device 100 according to the embodiment generates a model for predicting the delivery amount in the predetermined region based on the past delivery amount in the predetermined region and the history of the past user behavior information. 132 is further provided. The predicting unit 133 predicts the delivery amount generated in the predetermined area by inputting the behavior information acquired by the acquiring unit 131 to the model generated by the generating unit 132.

  As described above, the prediction apparatus 100 according to the embodiment can predict the delivery amount with high accuracy by generating a prediction model based on the past history and performing the prediction using the model.

  The embodiments of the present application and some of the modifications of the embodiments have been described in detail with reference to the drawings. However, these are examples, and the knowledge of those skilled in the art including the aspects described in the disclosure section of the invention is included. The present invention can be implemented in other forms based on various modifications and improvements.

  Further, the prediction apparatus 100 described above may be realized by a plurality of server computers, and depending on the function, an external platform or the like may be realized by calling an API (Application Programming Interface) or network computing. Can be changed flexibly.

  Further, “section (module, unit)” described in the claims can be read as “means”, “circuit”, and the like. For example, the generation unit can be read as generation means or a generation circuit.

DESCRIPTION OF SYMBOLS 1 Prediction system 10 User terminal 30 Web server 50 Delivery company apparatus 100 Prediction apparatus 110 Communication part 120 Storage part 121 Delivery history storage part 122 Model storage part 123 Behavior information storage part 124 Provision information storage part 130 Control part 131 Acquisition part 132 Generation part 133 Prediction unit 134 Providing unit

Claims (19)

Action information on the user's network , obtaining action information associated with a predetermined area, and storing the acquired information in a storage unit ;
The behavior information stored in the storage unit is input to a model that predicts the delivery amount in the predetermined area based on the relationship between the past delivery amount in the predetermined area and the history of behavior information on the user's network in the past. A prediction unit that predicts a delivery amount of a product that is a target of a commercial transaction and that occurs in the predetermined area;
A prediction apparatus comprising: The acquisition unit
As the behavior information, it is an order for a provider that provides the product, and acquires information related to an order with the predetermined area as a shipping destination,
The prediction unit
Predicting the amount of delivery occurring in the predetermined area based on the information about the order;
The prediction apparatus according to claim 1. The acquisition unit
As the behavior information, a search behavior related to the product, and information related to a search behavior executed by a user located in the predetermined area,
The prediction unit
Predicting a delivery amount occurring in the predetermined area based on information on the search behavior;
The prediction apparatus according to claim 1 or 2, wherein The acquisition unit
As the information related to the search behavior, obtain an increase rate of the number of searches using a keyword related to the product as a query,
The prediction unit
Predicting the amount of delivery occurring in the predetermined area based on the increase rate of the number of searches;
The prediction apparatus according to claim 3. The acquisition unit
As the behavior information, it is a post related to the product, and acquires information related to a post from a user located in the predetermined area,
The prediction unit
Predicting the amount of delivery occurring in the predetermined area based on the information about the posting;
The prediction apparatus according to any one of claims 1 to 4, wherein the prediction apparatus includes: The acquisition unit
As the information regarding the posting, at least one of a message created by the user, a text posted by the user to SNS (Social Networking Service), an email created by the user, and a review posted by the user on the platform that is the place of the commercial transaction. Acquired,
The prediction unit
A keyword included in the information acquired by the acquisition unit, and predicting a delivery amount generated in the predetermined area based on the number of appearances or the appearance rate of a keyword related to the product;
The prediction apparatus according to claim 5. The acquisition unit
Obtaining news information about the product in the predetermined area;
The prediction unit
Predicting a delivery amount occurring in the predetermined area based on the news report information;
The prediction apparatus according to claim 1, wherein A providing unit that provides a delivery amount predicted by the prediction unit to a delivery company that delivers the product in the predetermined area;
The prediction apparatus according to claim 1, further comprising: The providing unit includes:
Based on the delivery amount predicted by the prediction unit, the delivery company provides information on delivery resources provided in the predetermined area.
The prediction apparatus according to claim 8. The acquisition unit
Get action information including day of the week information or weather information,
The prediction unit
Predicting a delivery amount generated in the predetermined area based on behavior information including day information or weather information acquired by the acquisition unit;
The prediction apparatus according to claim 1, wherein Wherein a previous shipping amount in the predetermined region, based on the behavior history information of past user, generating unit that generates the model for predicting the shipping amount in the predetermined region,
Further comprising
The prediction unit
Predicting a delivery amount generated in the predetermined region by inputting the behavior information acquired by the acquisition unit into the model generated by the generation unit;
The prediction apparatus according to claim 1, wherein A prediction method performed by a computer,
Action information on the user's network , acquiring action information associated with a predetermined area, and storing the acquired information in a storage unit ;
The behavior information stored in the storage unit is input to a model that predicts the delivery amount in the predetermined area based on the relationship between the past delivery amount in the predetermined area and the history of behavior information on the user's network in the past. A predicting step of predicting a delivery amount of a product subject to a commercial transaction and occurring in the predetermined area;
The prediction method characterized by including. Action information on a user's network , acquiring action information associated with a predetermined area, and acquiring the acquired information in a storage unit ;
The behavior information stored in the storage unit is input to a model that predicts the delivery amount in the predetermined area based on the relationship between the past delivery amount in the predetermined area and the history of behavior information on the user's network in the past. A prediction procedure for predicting a delivery amount of a product that is a target of a commercial transaction and occurring in the predetermined area;
A prediction program that causes a computer to execute. An acquisition unit that acquires user behavior information associated with a predetermined area and stores the acquired information in a storage unit;
By inputting the behavior information stored in the storage unit into a model that predicts the delivery amount in the predetermined region based on the relationship between the past delivery amount in the predetermined region and the history of past user behavior information. A forecasting unit for forecasting the delivery amount of the product subject to the commercial transaction and occurring in the predetermined area;
With
The acquisition unit
As the behavior information, a search behavior related to the product, and information related to a search behavior executed by a user located in the predetermined area,
The prediction unit
Predicting a delivery amount occurring in the predetermined area based on information on the search behavior;
A prediction apparatus characterized by that. A prediction method performed by a computer,
An acquisition step of acquiring behavior information associated with a predetermined area, which is user behavior information, and storing the acquired information in a storage unit;
By inputting the behavior information stored in the storage unit into a model that predicts the delivery amount in the predetermined region based on the relationship between the past delivery amount in the predetermined region and the history of past user behavior information. A forecasting step for forecasting the delivery amount of the product subject to the commercial transaction and occurring in the predetermined area;
Including
The acquisition step includes
As the behavior information, a search behavior related to the product, and information related to a search behavior executed by a user located in the predetermined area,
The prediction step includes
Predicting a delivery amount occurring in the predetermined area based on information on the search behavior;
A prediction method characterized by that. An acquisition procedure for acquiring user behavior information associated with a predetermined area and storing the acquired information in a storage unit;
By inputting the behavior information stored in the storage unit into a model that predicts the delivery amount in the predetermined region based on the relationship between the past delivery amount in the predetermined region and the history of past user behavior information. A forecasting procedure for forecasting a delivery amount of a product subject to a commercial transaction and occurring in the predetermined area;
To the computer,
The acquisition procedure is as follows:
As the behavior information, a search behavior related to the product, and information related to a search behavior executed by a user located in the predetermined area,
The prediction procedure is:
Predicting a delivery amount occurring in the predetermined area based on information on the search behavior;
A prediction program characterized by that. An acquisition unit that acquires user behavior information associated with a predetermined area and stores the acquired information in a storage unit;
By inputting the behavior information stored in the storage unit into a model that predicts the delivery amount in the predetermined region based on the relationship between the past delivery amount in the predetermined region and the history of past user behavior information. A forecasting unit for forecasting the delivery amount of the product subject to the commercial transaction and occurring in the predetermined area;
With
The acquisition unit
As the behavior information, it is a post related to the product, and acquires information related to a post from a user located in the predetermined area,
The prediction unit
Predicting the amount of delivery occurring in the predetermined area based on the information about the posting;
A prediction apparatus characterized by that. A prediction method performed by a computer,
An acquisition step of acquiring behavior information associated with a predetermined area, which is user behavior information, and storing the acquired information in a storage unit;
By inputting the behavior information stored in the storage unit into a model that predicts the delivery amount in the predetermined region based on the relationship between the past delivery amount in the predetermined region and the history of past user behavior information. A forecasting step for forecasting the delivery amount of the product subject to the commercial transaction and occurring in the predetermined area;
Including
The acquisition step includes
As the behavior information, it is a post related to the product, and acquires information related to a post from a user located in the predetermined area,
The prediction step includes
Predicting the amount of delivery occurring in the predetermined area based on the information about the posting;
A prediction method characterized by that. An acquisition procedure for acquiring user behavior information associated with a predetermined area and storing the acquired information in a storage unit;
By inputting the behavior information stored in the storage unit into a model that predicts the delivery amount in the predetermined region based on the relationship between the past delivery amount in the predetermined region and the history of past user behavior information. A forecasting procedure for forecasting a delivery amount of a product subject to a commercial transaction and occurring in the predetermined area;
To the computer,
The acquisition procedure is as follows:
As the behavior information, it is a post related to the product, and acquires information related to a post from a user located in the predetermined area,
The prediction procedure is:
Predicting the amount of delivery occurring in the predetermined area based on the information about the posting;
A prediction program characterized by that.

Images