JP2001325495A - Purchase supporting device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a purchase supporting device capable of providing the information of high utility value to a client when the client is to purchase daily necessities such as perishable foods. SOLUTION: This device is provided with an input means for inputting attribute information containing information on the price of an article and a selling store related to the purchase of the article, a betting means 10 for handling the attribute information as betting to a database 15, an attribute information storage means 11 for storing the attribute information of betting in the database 15, classifying the information for each attribute value and generating the reliability of betting on the basis of the betting history in the past for each device or person inputting each of attribute information, a voting processing means 14 for extracting the information from the database as voting according to statistical processing, with which the generated reliability is considered, and a voting reporting means for reporting the extracted voting to a user. At the same time, the information of the price and the recommended store classified by one of categories, kinds, blocks and stores of the article is outputted as voting.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a purchase support apparatus capable of providing information such as a price that is useful when a customer purchases a product.

[0002]

2. Description of the Related Art With the development of a network environment in recent years,
At home, information such as product prices can be obtained via a network. For example, advertisements of estimated prices of various commodities are provided for each seller as a web service on the Internet.

[0003] Therefore, when a customer purchases a product, a plurality of sellers are presented on the Internet in order to identify the seller that sells the lowest price among the companies that provide advertisements as a WEB service. It is possible to search and collect the quoted prices to be made, and compare the prices of the respective sellers. A conventional technique for using price information existing on a network for purchasing goods at home is disclosed in, for example,
-63725. This conventional technology proposes that information such as a normal price / sale period of a product handled by a plurality of stores, an address of each store / business hours, and the like be collected from a network to support efficient purchase. .

[0004]

However, not all information required by customers is available on the network. Also, the information obtained on the network may not be useful to the customer. Specifically, there are the following problems. -Not all seller prices are offered on the network.

[0005] There is no database of price information of commodities across a plurality of sellers.・ When using barcode information as a database,
Since the content of the barcode information is not unique and is changed for each maker, each seller, and each season, it is necessary to respond to the change. -When obtaining information from barcodes, information on stores that do not employ barcodes cannot be collected.

[0006] The contents of the database are not always reliable. For example, there is a possibility that the information is manipulated by arbitrary trimming by the database creator. -It is difficult to reflect information on bargains, morning markets, time services, close-out sales, and other special sales that occur in each district, especially information that is not presented in flyers and the like. The above-described problem is unlikely to occur in appliances and the like that are to be mail-ordered on a network, but is likely to occur when purchasing daily necessities such as fresh foods.

[0007] It is an object of the present invention to provide a purchase support apparatus capable of providing high-value information to a customer when the customer purchases daily necessities such as fresh food.

[0008]

According to the present invention, each customer carries out voting on product prices and the like, and constructs a database of prices by seller, by district, by season, and the like. Each user who purchases a product shares and uses the knowledge of the market price that can be referred to at the time of purchase with the database. Such an environment is provided by the purchase support device of the present invention.

That is, according to the first aspect of the present invention, there is provided a purchase support apparatus comprising: input means for inputting attribute information including at least a price of a product related to product purchase and information of a store; Voting means for treating as voting to the database, and attribute information voted by the voting means are stored in the database and classified by attribute value, and each attribute information is input to each device or input user. Attribute information accumulating means for generating a voting reliability based on the voting history, voting processing means for extracting information from the database as voting in accordance with statistical processing in consideration of the generated reliability, and voting processing means. In addition to providing voting notification means for notifying users of the extracted voting, the product category, product type, district,
Information on prices and recommended stores classified by at least one for each store is output as the vote.

[0010] In claim 1, each customer uses the input means to input, for example, attribute information (price, store name, etc.) relating to the actually purchased product based on the display of the purchase receipt. This attribute information is treated as a vote on the database by the voting means. The attribute information storage means stores the attribute information voted by the voting means in a database and classifies the attribute information for each attribute value. In addition, the reliability of voting is generated for each device or each person who has input each attribute information based on the past voting history.

[0011] When a customer who intends to purchase a product refers to the information stored in the database, the vote processing means performs the statistical processing in consideration of the generated reliability.
Information is extracted from the database as a vote. This vote is notified to the user by the vote notification means. The content of the vote includes prices and recommended store information classified by at least one of product category, product type, district, and store.

When using such a database,
The price of a product entered as a vote is not always reliable. For example, there is a possibility of an error in displaying a price tag or the like attached to an actual product, an error in reading the price of an attribute information input user, an error in an attribute information input user input, or the like. Also, the reliability such as the price of a product input as a vote differs for each input person.

[0013] Therefore, if all the information input as a vote is trusted, highly reliable information cannot be obtained from the database. According to the first aspect, since the attribute information accumulating means generates the reliability of voting based on the past voting history, it is possible to distinguish between highly reliable voting and low-reliable voting. In addition, the voting processing means extracts information from the database as a vote according to the statistical processing in consideration of the reliability, so that when the user needs highly reliable information, only the highly reliable information is determined as a vote. Can be obtained.

According to a second aspect of the present invention, in the purchase supporting apparatus of the first aspect, for each vote from the input means, the input frequency, the difference between the input value and the displayed value of the vote, the input time, and the vote are used. At least one of product purchase status and erroneous input history
And calculating means for calculating an evaluation value relating to the use value by using the evaluation value, and using the evaluation value in a statistical process for determining the vote.

In the second aspect, an evaluation value relating to the use value is calculated by the calculation means, and this evaluation value is used in the statistical processing for determining the vote. Therefore, information with higher utility value can be extracted as a vote. For example, if the number of times that the user was actually notified as a vote vote is large, and the user actually purchased the product using the notification, the value of the information is high. Similarly,
For example, in the case of an input relating to a product on a special sale day, if the entered time is night, the product cannot be purchased at a special sale price, so the utility value should be regarded as low. Is considered high.

According to a third aspect of the present invention, in the purchase support apparatus according to the first aspect, at least for each device or each user who displays the vote extracted by the vote processing means, at least each user is evaluated by the impression information. An individual information storage unit is provided, and the impression information is used in statistical processing for determining the vote. Claim 3
In, for each user who uses the database,
The user's evaluation for each input user is registered, and this evaluation can be used for deciding a vote. Therefore, votes evaluated under different conditions for each user can be obtained.

Since the evaluation of each user is likely to reflect the preference of the product and the tendency of purchase, it cannot be used as a common evaluation value for voting by all users. According to the third aspect, a vote can be obtained by reflecting the evaluation of each user separately from the evaluation common to all users. Therefore, a vote reflecting the preference of the product and the tendency of purchase for each user can be obtained. can get.

For example, by increasing the evaluation of each user who provides information with high use value and lowering the evaluation of input users who provide information with low use value, The likelihood of obtaining a favorable vote increases. According to a fourth aspect of the present invention, there is provided the purchase support apparatus according to the first aspect, wherein a predicted purchase time of the product is estimated based on at least one of an input cycle related to input of attribute information associated with the purchase of the product and information on a shelf life of the product. Estimating means is provided, and a vote on a product to be purchased is automatically notified to a user at a product purchase time estimated by the purchase time estimating means or at a time in the vicinity thereof.

For example, commodities that are consumed on a daily basis must be purchased almost regularly and repeatedly so that shortages do not occur. Moreover, it is necessary to immediately purchase a product whose expiration date has passed even if the remaining amount is sufficient. However, it is common for consumers to forget to buy a product that needs to be purchased. According to the fourth aspect of the present invention, at the expected product purchase time or at a time near the expected time, a vote on the product to be purchased is automatically notified to the user, which helps prevent forgetting to buy.

According to a fifth aspect of the present invention, there is provided the purchase support apparatus according to the fourth aspect, wherein a buying time at which the price of the product obtained as the vote is reduced is detected in the vicinity of the product purchasing time estimated by the purchasing time estimating means. Detecting means is provided, and at the time corresponding to the time of the purchase detected by the buying time detecting means, a vote on the product to be purchased is automatically notified to the user.

For example, there are many stores that sell at low prices only on specific days of the week. If a product can be purchased only on such a special sale day, the product can be purchased at a lower price. In claim 5, the buying time detecting means detects the buying time when the price of the product obtained as a vote decreases based on the contents of the database. At the time corresponding to this purchase, the vote on the product to be purchased is automatically notified to the user, so the user recognizes that it is necessary to purchase when the product is inexpensive, such as a special sale date, and You can buy goods.

According to a sixth aspect of the present invention, in the purchase support apparatus of the first aspect, there is provided a moving cost calculating means for calculating a moving cost according to a positional relationship between a position where the user is present and a position where the product is sold. The statistical processing for deciding a vote is characterized by evaluating the information on the travel cost.

When purchasing a product at a store located far from the current location, the cost of transportation (transportation fare, fuel cost, etc.) cannot be ignored. If there is a distant store where products can be purchased at a lower price, purchasing at a nearby store may result in lower prices. According to the sixth aspect, the vote is determined by evaluating the information of the moving cost determined by the positional relationship, so that the user can obtain the information of the commodities that are cheaper overall as the vote.

According to a seventh aspect of the present invention, there is provided the purchase support apparatus according to the first aspect, further comprising inventory management means for storing information on a stock amount for each product and a storable capacity for each product on the user side, and It is characterized in that a purchaseable amount determined by the capacity is added to a vote notified to the user. In claim 7, since the user is notified of the purchaseable amount as a vote, it is useful to suppress the user from purchasing an unnecessarily large amount of commodities.

According to an eighth aspect of the present invention, there is provided the purchase support apparatus according to the first aspect, wherein a product having an attribute indicating purchase of a product and an attribute indicating the purchase time among the product information stored in the database for a specific period of time. A cash management means for notifying a user of a list of all the products purchased within the customer as the vote. In claim 8, the user is notified of a list of all the products purchased within a specific period as a vote, so that this device can be used as a function of a household account book.

[0026]

DESCRIPTION OF THE PREFERRED EMBODIMENTS (First Embodiment) One embodiment of a purchase support apparatus according to the present invention will be described with reference to FIGS. This embodiment corresponds to claims 1 to 3. FIG. 1 is a block diagram showing a functional configuration of the purchase support device of this embodiment. FIG. 2 is a flowchart showing a specific example of the information input process of this embodiment. FIG. 3 is a flowchart showing a specific example of the input receiving process of this embodiment. FIG. 4 is a flowchart showing a specific example of the customization processing of this embodiment. FIG. 5 is a flowchart showing a specific example of the impression input processing of this embodiment. FIG. 6 is a flowchart showing a specific example of the search processing of this embodiment.

FIG. 7 is a block diagram showing a hardware configuration example (1) of the purchase support apparatus. FIG. 8 is a block diagram showing a configuration example (2) of the hardware of the purchase support device. FIG. 9 shows a hardware configuration example (3) of the purchase support device.
FIG. In this embodiment, the input means, the database, the voting means, the attribute information storage means,
The voting processing means and the voting notification means are respectively provided by the input device 2
1, a storage device 15, an information input process 10, an input reception process 11, a search process 14, and a display device 23. In addition, the calculating means of claim 2 performs step S55.
The individual information storage means of claim 3 is embodied as an individual impression information DB (abbreviation of database: the same applies hereinafter) 154.

The purchase support apparatus of this embodiment includes, for example, an input device 21, an information processing device 22, a display device 23, and a storage device 15, as shown in FIG. Information processing device 22
Constructs a database on the storage device 15 according to information input (voting) from the input device 21. Further, the information processing device 22 searches the database on the storage device 15 according to the search instruction from the input device 21, and outputs the result to the display device 23.

The input device 21 is realized by one or a combination of an image scanner, a bar code reader, a television camera, a numeric keypad, a keyboard, and a character recognition device. The display device 23 is a small portable device such as a liquid crystal display device, and has at least a function of displaying numbers and character strings. Input device 21 and display device 2
3 may be integrally configured as, for example, the customer terminal 20 shown in FIG. 8, or may be configured as an independent device.

In the example of FIG. 8, a plurality of customer terminals 20 are connected to a service center 30 via a communication network. As a communication network, for example, a mobile communication network, a dedicated line, the Internet, or the like may be used. In the example of FIG. 8, the information processing device 22 and the storage device 15 are provided in the service center 30. Further, in the example of FIG.
The information processing device 22 and the storage device 15 are provided in (4).

In FIG. 9, the information processing device 22 and the storage device 15 are installed in a single customer terminal. However, the information processing device 22 and the storage device 15 are installed in a plurality of terminals and function independently or cooperatively operate with each other. A form that functions as a single device is also possible. Therefore, it is not necessary to provide a special service center. In the example of FIG. 9, the customer terminal 20 (4) and the customer terminal 20
Communication is performed between (1) to (3).

The input device 21 and the display device 2 shown in FIG.
3 and the customer terminals 20 in FIGS. 8 and 9 need to be provided in one or more sets. Further, more customer terminals 20 may be connected to the purchase support device than in the example of FIG. Further, a part of the functions of the information processing device 22 and the storage device 15 may be provided in each customer terminal 20. As shown in FIG. 1, the purchase support device of this embodiment has functions of an information input process 10, an input reception process 11, a customization process 12, an impression input process 13, and a search process 14. These functions are realized by the information processing device 22 shown in FIGS.

Here, the input of the information for constructing the database on the storage device 15 of the purchase support apparatus is performed by inputting the customer (general consumer) who wants to purchase the product, the customer who actually purchased the product, and the information. It is assumed that it is performed by the providing customer. In addition, users who try to obtain information (information on inexpensive products) useful for purchasing products using the database of the purchase support device are also ordinary consumers. However, the user is not necessarily the same as the input user who inputs the information for constructing the database.

The information that each customer inputs from the input device 21 is attribute information of a product related to the purchase of the product or a store that sells the product. The content of this attribute information is
Purchase store name, purchase store location or area, product name or product type, product price, purchase date and time, sales unit (how many sets, how many bundles,
Kg, etc.), discount information of the store that sells the product itself can be considered.

When discounts and discounts are considered, it is desirable to include in the attribute information the presence or absence of discounts and discounts, the reason, conditions for receiving discounts, and the like. Furthermore,
When considering the upper limit of bulk purchases per person or at one time, it is desirable to include the presence or absence and the upper limit in the attribute information, and if there is an evaluation regarding the quality of the product, the evaluation should also be included in the attribute information. desirable.

It should be noted that the purchase date and time can be replaced with the input date and time unless the purchase date and time are important for bargain or time service, for example. Reasons for discounts include bargains, time services, use of coupons,
Discount sales, close-out products, etc. are possible. If the reason for the discount is a bargain or time service, it is desirable to include information on the end time of the bargain or the time zone of the permanent time service in the attribute information.

When the reason for the discount is the use of a coupon, it is desirable to include information about ease of use, such as whether the coupon is available to anyone at the store or a coupon limited to a special customer. In the case of discount by age, it is desirable to include information on the target age, and when the number of sales is limited, it is desirable to include the information of the current constant. Such product attribute information can be input in various ways. For example, when an image scanner is provided as the input device 21, various information of a product can be recognized from an image obtained by reading a purchase receipt with the image scanner.
When a bar code reader is provided as the input device 21, the bar code attached to the product is read by the bar code reader so that the attribute information on the product can be recognized. Further, it is also possible to input attribute information of the product manually by using a numeric keypad, or to identify the product from an image read by a television camera.

When each customer inputs attribute information of a product, the information input process 10 of FIG. 1 is executed, and the result is input to the input receiving process 11 as a vote. For example,
When reading information from a purchase receipt as shown in FIG. 13 using an image scanner, a purchase receipt reading process 10A shown in FIG. By reading the purchase receipt, the name of the store, the location of the store, the name of the store, the amount, the price, the number, the reason for the discount, the amount of the discount, and other descriptions useful for purchase support (for example, fresh food 5 % Discount).

A description will be given with reference to FIG. In the first step S11, an image of a purchase receipt is read using an image scanner. In step S12, step S11
Processes the image read by and performs character recognition. However, there are cases where it is difficult to recognize characters, and cases where part of the information to be input is not described in the purchase receipt. In order to support the input operation in such a case, estimation and complementation are performed in the next step S13.

That is, the past input history of the input user is checked, or a database (in this case, the product information DB 15
By examining similar information registered in 1), it is possible to estimate and supplement character strings that are difficult to recognize or contents of insufficient information. When there are a large number of input devices 21, the input person can be distinguished by inputting a number specifying one input device 21 or an identification number assigned to the input person.

For example, if the name of the input attribute information is meat,
When the price is 100 g and 100 yen, as similar information on the database, if the input of a similar amount (including the input of another customer) is “pork rose” at the same store in the same store, “ Estimate and complement "pig roses". In addition, if the input person has a larger purchase history of thigh meat than other meats, and the input amount is within a reasonable amount of “thigh” estimated by statistical processing, The type is assumed to be "thigh" and complemented.

If, for example, the discount is simply described for the discount and the reason is unknown, the following processing is performed. For example, if the purchase time is 10 minutes before closing of the purchase store, it is regarded as “discontinuation item discount”. In addition, if the input is related to a store that has a discount by age and is an input from an input person who has a history of frequently using age discount, it is estimated to be “age discount”.

However, the information supplemented in step S13 may be incorrect. Therefore, in step S14, information (such as an image) read from the purchase receipt and estimated information are displayed on the display device 23, and confirmation of the input person is requested. When there are a plurality of candidates as the estimation information, those options are displayed on the display device 23 in step S14. In step S15, one of the options is specified as correct information by the selection input of the input person whose contents have been confirmed, or correct information is newly input manually.

When a bar code reader is provided as the input device 21 and a bar code is added to the product, the lack information can be complemented by using the bar code reading together. When reading a barcode uniquely assigned by a manufacturer that manufactures a product, it is possible to read the item of the product and the normal price. However, in that case, the actual purchase price is unknown. In that case, the purchase price can be estimated by performing statistical processing on the information on the products already registered in the database and the list price.
Therefore, the correct purchase price is specified by displaying the estimated purchase price options, the purchase price is manually input, or the correct purchase price is specified by associating the estimated purchase price with the price read from the receipt. .

In the case where a bar code assigned to each season or each store by a distributor is read, and the name of the purchase store is uniquely identified by reading the receipt, the information of the purchase store is searched from the database. Then, barcode information of the purchase store is searched, and attribute information of the product is estimated from the search result. Also in this case, the estimated information is displayed, and the correct information is specified after confirming the input person.

A specific example of an example of confirmation regarding a product item will be described below. If “special sale item, domestic product, pork, roasted meat, 100 g” is obtained as a result of purchase receipt reading and barcode reading, “fresh food”
"Meat", "pig", "domestic", "rose", "100g",
The respective attributes of the "special sale item" are displayed, and the portion of the attribute that is incorrect is specified by the input of the input person.

For example, by reading the purchase receipt and barcode of the product actually purchased, (1) fresh, meat, pork, domestic, rose, 50 g, ordinary product, (2) fresh, meat, pig, import, rose, 100g, sale item (3) Fresh, meat, pork, domestic, foot, one pair, normal (4) fresh, meat, pig, domestic, rose, 50g, normal (5) fresh, fish, horse mackerel,, open (3) Information such as decorations, daruma,,,, one, and special sale items can be obtained, and these are displayed on the display device 23.

Here, for example, when "pig" in (4) is an error of "cow", "pig" in (4) is selected as a correction target by the selection input of the input person. In that case, the choices are "cow?", "Horse?", "Other meat?", "Fresh food other than meat?", "Other than fresh food?", "Other than food?" indicate. Here, the input person is "cow?"
Is selected, the attribute information of (4) is modified as follows.

(4) Fresh, Meat, Beef, Domestic, Rose, 50g, Normal Product Also, for example, if “pork” in (4) is an error of “Aji”, before correcting the error of “meat” If you select "Pig" for
The correct "Aji" is not a "meat" but a "fish", so "Aji" is not displayed as a correction candidate. In such a case,
By selecting "Fresh food other than meat?", Correction to "Aji" becomes possible. Here, when the complementation by the input in S15 of FIG. 2 is not performed, the input is treated as an incomplete voting.

Also, when the attribute information of the commodity is input by reading the bar code from the beginning without reading the receipt, the input is performed by the same processing as described above. That is, when reading a barcode assigned to each season or each store by a distributor, the information for specifying the store to be purchased is manually input, or based on the information of the store estimated from the past input history. Identify the contents of the barcode. Then, the estimated purchase price obtained by statistically processing the read list price and the information on the database is displayed, and the correct purchase price is specified by selection input or direct manual input by an input person.

Further, the attribute information may be directly input manually by the input person without performing the receipt reading or the bar code reading. However, in that case, it is desirable to display input candidate options such as “fresh food?”, “Frozen food?”, And “confectionery?” To support the input. As a result, even when a simple input device such as a numeric keypad is used, information can be input with a small burden.

Information input by the information input process 10 as described above is registered in the storage device 15 via the input receiving process 11 of FIG. A specific example of the input receiving process 11 will be described with reference to FIG. Step S2
0, the product information DB 15 configured on the storage device 15
With reference to the content of No. 1, information on products of the same classification (similar products) is searched.

In the product information DB 151, based on inputs (votes) from various input users (customers and consumers of products),
For each of the various products, the identification number of the input person (or the number of the input device), the name of the purchase store, the location (or area) of the purchase store, the product name (or product type), the product price (purchase price), the purchase date and time ( Or input date and time), sales unit, and the like are registered.

The product information DB as a result of step S20
By statistically processing a large number of pieces of information on the same kind of product obtained from 151, it is possible to specify a normal price range of the product. In step S21, the price input this time as a vote from the input person is compared with a normal price range obtained by statistical processing of information extracted from the product information DB 151. If the input price is within the appropriate range, the process proceeds from step S22 to S27, and if it is an abnormal value, the process proceeds from step S22 to S23.

In step S23, it is confirmed whether or not there is a reason for the abnormal value. For example, when a special reason such as a bargain is added to the input information, it can be determined that even an abnormal value is a reasonable price. Therefore, if there is a reason, steps S23 to S2
Then, the process proceeds to S27 through 4. In the case of an abnormal value without a reason, the process proceeds from step S23 to S25 through S24.

In step S25, the product information DB 151
Is searched, and the input price from another input person at the same store of the same product is checked. Even if rational reasons such as discounts are unknown, if multiple entrants (for example, three) enter prices close to each other, the price may be considered to be statistically normal. it can. That is, it can be considered that an effective discount or price revision has occurred in the store.

Therefore, when a predetermined number of inputters are inputting prices close to each other, the input price is regarded as a normal value, and the process proceeds from step S25 to S26 to S27.
Proceed to. If the input price cannot be regarded as a normal value, the process proceeds from step S25 to S28 through S26.
In step S27, the normal input history of the input person is stored in the trust information DB 152. In step S28, the history of the erroneous input of the input person is stored in the trust information DB 152.

In step S29, all of the input attribute information is used as a new vote from the input person in the product information DB1.
51 is additionally stored. By executing the input receiving process 11 of FIG. 3, not only the product attribute information is stored in the product information DB 151, but also the input history regarding the reliability of each input person is stored in the trust information DB 152.

That is, many "normal input" histories are accumulated for a reliable input user who inputs a correct price, and many "erroneous input" histories are accumulated for an unreliable input user who inputs an incorrect price. Is done. Therefore, by referring to the contents of the trust information DB 152, it is possible to check the reliability of each input user of the information stored in the product information DB 151.

Further, in this example, conditions such as various histories registered in the custom information DB 153 of FIG. 4 can be reflected in search conditions for extracting information from a database (such as the product information DB 151) as a vote. The custom information DB 153 is provided individually for each user who uses the information of the database constructed on the storage device 15, so that each user stores information for reflecting his / her preference and ease of use. Custom information D assigned to a person
B153.

The custom information DB 153 includes an individual impression information DB 154, an individual input history DB 155, and a user input history DB 156. In addition, custom information DB
153 may be provided on the storage device 15 or on the customer terminal 20 owned by each user. When the user performs an input for customization, the customization processing 12 in FIG. 4 is executed. The contents of the processing in FIG. 4 will be described below.

In the customization process 12, the information of the impression of each user with respect to another customer (the person who inputs information) can be registered in the individual impression information DB 154. For example, the customer's vote is often wrong, or the customer's preference is unreliable. The information output as a vote by the search for the purchase support device includes an identification number for identifying the input person of the information or the number of the device possessed by the input person. Therefore,
The user who uses the information of the vote can display the number of the input person who provided the information and can specify the person.

Therefore, each user registers the evaluation of the information such as the commodity price output as a vote by the search of the purchase support device as an impression on the input person who provided the information. When the user inputs the impression information and the number of the corresponding person, the process proceeds from step S31 to S32 in FIG. 4 to update the impression information for the corresponding person on the individual impression information DB 154 with the impression information input this time.

If the information output as a vote by the search of the purchase support apparatus is found to be incorrect, the user inputs information for pointing out the error along with the number of the person concerned. In that case, the process of the purchase support device proceeds from step S33 to S34. That is, the history of the erroneous input of the corresponding person is added to the individual input history DB 155.

When the user finds that the information output as a vote by the search of the purchase support apparatus is useful to the user himself, the user inputs information for pointing out the information together with the number of the person concerned. I do. In that case, the process of the purchase support device proceeds from step S35 to S36. That is, the history of the useful input of the corresponding person is stored in the individual input history DB 155.
Is added to

However, the history of the erroneous input in step S34 and the useful input in step S36 is a result of the judgment of a specific user, and is not always available to all users. Therefore, these histories are arranged only in the custom information DB 153 of a specific user, and are not reflected in the search conditions for voting by other users.

When the user obtains information useful for purchasing a product as a vote from the database of the purchase support apparatus, it is necessary to specify various search conditions. When the search condition is input by the user, the process of the purchase support device proceeds from step S37 to S38. That is, the history of the search condition input this time is added to the user input history DB 156.

By using the history on the user input history DB 156, a search can be performed under the same search conditions as those used previously, or a search with high frequency can be found from a large number of past histories and this search can be performed. It can be used as a condition candidate. For example, it is possible to presume from a history of search conditions that the user prefers a store convenient for purchasing a product using a car. It is also possible to estimate whether there is a tendency to collect miles and coupons. Further, it is also possible to estimate from the history the price difference of the product as a criterion for selecting the lowest price store and the store that can collect miles and coupons.

Further, on the premise that a plurality of products are purchased, it is also possible to check from the history whether or not there is a tendency to select a store where the total product price is the lowest. It is also possible to group a plurality of stores having similar access tendency from the history into the same group and search for the lowest price store for each group. In the customization processing 12 of FIG. 4, the input impression information is reflected only in the search condition of the vote used by the individual user, but the impression information considered to be statistically correct is common to the search of all users. It can be reflected as an evaluation value used for.

The following describes the impression input process 13 shown in FIG. When the user inputs an indication of an erroneous input, the process proceeds from step S41 to S42, and the history of the erroneous input of the user is added to the evaluation input history DB 157 together with the number of the user who has indicated the erroneous input.

If the user inputs an indication of a useful input, the process proceeds from step S41 to S45 and S46, where the history of the useful input of the relevant person is evaluated together with the number of the user who pointed out the evaluation input history. Add to DB157. In step S43, the contents of the evaluation input history DB 157 are searched, a history of all erroneous input indications is extracted for each input person (person who provided the information), and the number of users who have indicated those errors is determined in advance. Compare with the determined threshold (number of people that can be considered statistically correct).

When a user who is equal to or more than the threshold value points out an error, the process proceeds from step S43 to S44, and the evaluation value for the corresponding person (input person) is updated. This evaluation value is a numerical value representing the evaluation (reliability or not) seen from other users, and can be handled in the same manner as the reliability. The evaluation value for each input person is stored in the trust information DB 152. Similarly,
In step S47, the contents of the evaluation input history DB 157 are searched to extract the history of pointing out all useful inputs for each input person (the person who provided the information), and the number of users who pointed them out is determined in advance. Compare with threshold (number of people that can be considered statistically correct).

If the user who is equal to or more than the threshold value points out that it is useful, the process proceeds from step S47 to S48, and the evaluation value for the corresponding person (input person) is updated. That is, a low evaluation is given to an input user who points out an input error by a large number of users, and a high evaluation is given to an input user whose many users recognize a useful input.

When each user wants to obtain information (voting) useful for purchasing a product from the database of the purchase support apparatus, the search processing 14 shown in FIG. 6 is executed. The contents of this processing will be described below.

Each user who uses this purchase support apparatus,
You can specify a custom search when searching. FIG.
In step S51, it is checked whether or not a custom search has been designated.
Proceed to. In step S52, the history of the user input history DB 156 included in the custom information DB 153 assigned to the user is checked, and one or more frequently used search conditions are selected by the user as search conditions that can be selected. It is displayed on the display device 23 of the terminal 20. In step S53, one search condition is selected according to the user's input.

When the search condition is specified from the beginning or when a part of the search condition selected in step S53 is changed, the condition is input or changed in accordance with the user's input in step S54. In step S55, the trust information DB
152, individual impression information DB 154 and individual input history DB
The reliability and the evaluation value are calculated for each input person (provider of information) with reference to the history stored in 155.

For example, a history of 100 entries for a certain input person exists in the trust information DB 152, and 5
If there is a history in which “erroneous input” is detected in the number of times of input, the reliability is calculated as 95%. The same applies to the evaluation value. Further, the final reliability and the evaluation value are calculated so as to reflect the histories of the individual impression information DB 154 and the individual input history DB 155.

In step S56, the product information DB 151
And narrows down the search target using the reliability and the evaluation value of each input person calculated in step S55. In other words, product information entered as voting by an entrant who frequently provides incorrect information is excluded from search targets in advance,
Give priority to the information of the input person with high reliability. Step S57
Then, the search information is narrowed down according to the search conditions specified in step S54 and the like. For example, the product information of a store that sells a specific product at the lowest price is extracted from the product information DB 151 as a vote.

In step S58, the result of the search in step S57 is displayed on the display device 23 of the user as a vote. Here, the number of the input person who provided the information extracted as a vote may be displayed on the display device 23.
If a plurality of information is displayed as a vote in step S58, there is room for further narrowing down the plurality of votes. In that case, when the user adds a search condition, step S5
The process returns from S9 to S57 to narrow down the vote.

If the search is to be performed again, the process proceeds from step S59 to S51 via S60, and the above processing is repeated. The order of narrowing down the search may be changed as necessary. Voting means the result of statistical processing for confirming the reliability of voting for information on product purchases, narrowing down information required by users, and providing information on “recommended degree”. I have.

In practice, the possibility of erroneous input, which can be determined from the past input history of each voting customer, is used as reliability information and used for voting. For this reason, it is possible to prevent misreading at the time of purchase based on an incorrect input. In addition, even if a store misuses the purchase support device and intentionally inputs incorrect information in order to attract customers, it is possible to reduce the reliability of the information and prevent the information from appearing in votes. Misuse can be prevented.

By performing the search processing 14 shown in FIG. 6, for example, products are classified by category, product type, district, store, time (season, day of the week, special sale date), or a combination thereof. At the same time, it is possible to notify (display) a vote selected based on the price determined in the processing of the average value, the mode, the lowest price, and the like as information useful for purchasing the product.

For example, it is possible to present to the user as a vote at which store the designated product can be purchased at what price. In this case, incomplete voting that is a fish but it is not clear what kind of fish is, or voting information that requires a special discount that is unavailable to customers using voting is excluded from voting. For example, the following vote can be output as a search result for a single horse mackerel search.

[0084] Cheapest store: A series, B branch 100 yen with 20% reliability. Cheapest store with high reliability (Narrowed down with reliability): Reliability 9
9% for 105 yen at C store (only 100 limited). The cheapest shops that can be reached by car (access method narrowing down):
106 yen at D department store with 99% reliability.

In addition, narrowing down by an access range (for example, excluding the area outside the Tokyo area) and a limited number of sales (for example, limited to 10
Narrow down by 0, specify time (eg 1 in the evening)
It is also possible to output a vote as a result of narrowing down by 8:00). Further, a process may be performed in which information of a product output as a vote by the purchase support device is directly linked to the purchase of the product. In this case, if the seller sells online, connect directly to the seller online.If the seller does not sell online, connect to the purchase agent or delivery company online to purchase. It should be linked to the processing to be entrusted.

As described above, the price information that does not exist in the database provided by the seller online, such as WEB, and the information on the occurrence of occasional special sales are also converted into a database using the purchase support apparatus of the present invention. Can be used. In addition, when entering information on products equivalent to products already registered in the database as a vote,
Insufficient information is supplemented using information stored in the database, so that input of information is easy.

Further, since the reliability of the input person is determined, it is possible to correct an input error or intentional erroneous data input, thereby improving the reliability of the database. In addition,
The reliability and the evaluation value given to the voting from each input person can be calculated under any one or a combination of the following conditions. -Frequency of input.

The difference between the input value and the displayed value of the vote. -Input time. -Whether there is an input having the same attribute information after displaying the vote.・ Reliability based on the ratio of erroneous input.・ Validity of voting.

A ratio referenced by another customer. The purpose of such reliability and evaluation values is to provide an incentive for data input processing in addition to constructing a correct database. desirable. -Those with high use value.

[0098] No lie. -Inputs with many available attributes.・ Easy input time. As for the use value, for example, it is desirable that the number of times of display as a vote is large and that the evaluation is high when the purchase is executed by the display. Also, when the lowest price is displayed as a vote, the number of times the customer who inputs the lowest price displays the vote is increased, but the number of times the same price is input after the display can be regarded as the utility value.

For example, if the input time is a special sale date, if the input is at night, the special sale date has already ended, and it cannot be used to support the purchase on that day. The earlier you should be, the better. However, even in the case of night time input, by counting the number of times of input at the same price after the vote is displayed, it is possible to give meaning to confirm the value of the input information of another person.

Further, since it is meaningful as information for estimating the price of the next bargain sale day or estimating the bargain period, it is appropriate to assign an evaluation value to the input time. Because of the significance of the input at this early time, it is desirable to input the information of the flyer obtained early in the morning regardless of the purchase. As described above, an incentive can be given to an input by giving an evaluation value to a vote from each input person.

(Second Embodiment) Another embodiment of the purchase support apparatus of the present invention will be described with reference to FIG. This embodiment corresponds to claim 4. This embodiment is a modification of the first embodiment. This is the same as the first embodiment except that the automatic search process shown in FIG. 10 is added to the process of the information processing device 22. The following description is omitted for the same parts.

In this embodiment, the purchase time estimating means of claim 4 is embodied as steps S70, S71, and S73. The automatic search process shown in FIG. 10 is periodically and repeatedly executed. In the first step S65, the information of the current date and time is acquired using the calendar function built in the information processing device 22.

The input person input history DB 161 of FIG. 10 holds an input history of voting for each input person. When the information of the product is newly input as a vote from the input device 21, the process proceeds from step S66 to S67. In step S67, with respect to the voting history of the same input person, the input history relating to the same or similar product as the product input this time is searched from the input user input history DB 161, and the previous input date and time (the date and time of the last stored history) are retrieved. ) Is detected.

In step S68, the time difference between the current date and time obtained in step S65 and the previous input date and time detected in step S67 is calculated as the input cycle for the product. In step S69, the input history regarding the product input this time is added to the input user input history DB 161. The information registered as the history includes the input date and time, the product type, the expiration date, the input cycle, the identification number of the input user, and the like.

In step S70, for each user, the input history for each product is examined by searching the input user input history DB 161. In step S71, the expiration date is checked for each product. That is, the expiration date of the last registered product is compared with the current date and time acquired in step S65 to determine whether the expiration date is approaching. If the expiration date is approaching, step S71
Then, the process proceeds to S72, and a warning of the expiration date of the product is displayed on the display device 23.

In step S73, the expected shortage date and time for each product is confirmed. That is, a result obtained by adding the input cycle to the input date and time (previous purchase date and time) of the last registered product is detected as the expected shortage date and the expected shortage date and time are compared with the current date and time acquired in step S65. It is determined whether or not the shortage date and time when the product is out of stock is approaching. If the shortage date is approaching, the process proceeds to step S74, and a warning of a shortage prediction regarding the product is displayed on the display device 23.
To be displayed.

That is, in this embodiment, the shortage date and time expected from the input date and time and the input cycle of the same type of product input from the input device 21 and the expiration date of the product to which storage environment information such as the expiration date are added are added. Based on at least one,
A vote on a product that is expected to be purchased can be automatically displayed. With this function, it is possible to give an incentive to the customer to vote, warn of recognition of the expired product, and prevent the customer from inadvertently forgetting to buy.

Note that the prediction of the purchase time may be processed as follows. (1) The purchase cycle is predicted based on the frequency of voting input from the input device 21. (2) By additionally inputting the usage status (such as the stock amount) from the input device 21, a more accurate shortage time is predicted and determined. In addition, by using such storage information, a menu of meals that can be created with stock items is automatically proposed, a menu of meals with stock items is displayed, and purchase support for products that are lacking in the menus is provided. It is also possible to display information.

(Third Embodiment) Another embodiment of the purchase support apparatus of the present invention will be described with reference to FIG. This embodiment corresponds to claim 5. This embodiment is a modification of the second embodiment. This is the same as the second embodiment except that the content of the automatic search process is changed as shown in FIG. In FIG. 11, the description of the processing between steps S65 and S70 is omitted, but the omitted part is the same as in FIG. The following description is omitted for the same parts.

In this embodiment, the buying time detecting means in claim 5 is embodied as steps S75 and S77. In step S71B, the expiration date is checked for each product. However, instead of detecting the same timing as the expiration date, it is determined whether or not it is near the expiration date, for example, within one week before the expiration date. If the current date and time is near the expiration date, the process proceeds from step S71B to S75.

At step S75, the product information DB 151
And the time of purchase (for example, when the price becomes low) of a product whose expiration date is approaching is searched. In other words, by statistically examining the past input history of the customer, the time when the price of the product becomes low due to special sale or the like is detected as buying time, or the specific day of the week when the customer frequently performs shopping is considered as buying time. Can be detected.

If the current date and time is the time of purchase of a product whose expiration date is approaching, the process proceeds to step S72 via step S76, and a warning of the expiration date is displayed. Step S7
At 3B, the expected shortage date and time is confirmed for each product.
However, instead of detecting the timing of the expected shortage date and time itself, it is determined whether or not it is near the expected shortage date and time, for example, within one week before the expected shortage date and time. If the current date and time is within the period near the expected shortage date and time, the process proceeds from step S73B to S77.

In step S77, the product information DB 151
And the time of purchase (for example, when the price becomes low) of a product near the expected shortage date and time is searched. In other words, by statistically examining the past input history of the customer, the time when the price of the product becomes low due to special sale or the like is detected as buying time, or the specific day of the week when the customer frequently performs shopping is considered as buying time. Can be detected.

If the current date and time is the time of purchase of a product near the expected shortage date and time, the process proceeds to step S78 through step S78.
Proceed to 4 to display a warning of a shortage expectation. In this form,
Rather than simply linearly estimating the date and time at which the customer should purchase the product, it is possible to detect the day of the week or the special sale date when the price is lower as the time of purchase and recommend the customer to purchase at that timing. In addition, a more timely proposal can be made to a customer who purchases only on a specific day such as Sunday.

(Fourth Embodiment) Another embodiment of the purchase support apparatus of the present invention will be described with reference to FIG. This embodiment corresponds to claim 6. This embodiment is a modification of the first embodiment. It is the same as the first embodiment except that the content of the search processing 14 is changed as shown in FIG. In FIG. 12, the same steps as those in FIG. 6 are denoted by the same reference numerals. The following description is omitted for the same parts.

In this embodiment, the moving cost calculating means in claim 6 is embodied as step S82. FIG.
In step S81, information on the current position of the product purchaser (user) (or the position of the customer terminal 20) is input. In step S82, a travel cost determined by the positional relationship between the current position of the product purchaser and the sales position of the product (the location of the sales store) is calculated for each product information held in the product information DB 151.

In step S57B, a vote is generated by narrowing down the information of each product in consideration of the obtained moving cost. With this function, for example, in the case of a product such as gasoline that has a different price in each district, it is possible to display only the information in the district where it can be purchased. Also,
It is possible to provide sales support in consideration of the net cost required to obtain the product in addition to the travel cost incurred when using time and transportation.

Further, whether or not a commuter pass is possessed, whether or not a car is owned, whether parking fees are high or low, the positional relationship between parking lots and customers,
By taking into account distance information between the parking lot and the store, information about congestion on the road, shipping charges for delivery, and the like, it is possible to calculate a more accurate moving cost. (Fifth Embodiment) Another embodiment of the purchase support apparatus of the present invention will be described. This embodiment corresponds to claim 7.

Although not shown, in this embodiment, there is provided a stock management means for holding information on the stock amount of each product and the storable capacity of each product on the user side. In addition, a purchaseable amount determined by the stock amount and the storable capacity is added to a vote for notifying the user. In other words, since a list of purchaseable amounts for each product can be presented to the purchaser, it is possible to prevent the purchaser from over-purchasing the product and to provide a voting incentive.

(Sixth Embodiment) Another embodiment of the purchase support apparatus of the present invention will be described. This embodiment corresponds to claim 8. Although not shown, in this embodiment, a list of all the products purchased within a specific period is displayed for a product having an attribute indicating the purchase of the product and an attribute indicating the purchase time of the product information stored in the database. A cash management means for notifying the user as a vote is provided.
In this mode, since a list of purchased products can be displayed, the purchase support device can be used as a cashier,
And it can provide incentives for voting.

[0113]

As described above, according to the present invention, a database of various products can be constructed based on input of information from customers, and information useful for each customer's product purchase can be provided. .

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a functional configuration of a purchase support device according to a first embodiment.

FIG. 2 is a flowchart illustrating a specific example of an information input process according to the first embodiment;

FIG. 3 is a flowchart illustrating a specific example of an input receiving process according to the first embodiment;

FIG. 4 is a flowchart illustrating a specific example of a customization process according to the first embodiment.

FIG. 5 is a flowchart illustrating a specific example of an impression input process according to the first embodiment;

FIG. 6 is a flowchart illustrating a specific example of a search process according to the first embodiment.

FIG. 7 is a block diagram showing a configuration example (1) of hardware of a purchase support device.

FIG. 8 is a block diagram illustrating a hardware configuration example (2) of the purchase support apparatus.

FIG. 9 is a block diagram showing a configuration example (3) of the hardware of the purchase support apparatus.

FIG. 10 is a flowchart illustrating a specific example of an automatic search process according to the second embodiment.

FIG. 11 is a flowchart illustrating a specific example of an automatic search process according to the third embodiment.

FIG. 12 is a flowchart illustrating a specific example of a search process according to the fourth embodiment.

FIG. 13 is a plan view showing an example of a purchase receipt.

[Explanation of symbols]

 Reference Signs List 10 information input processing 10A purchase receipt reading processing 11 input receiving processing 12 customizing processing 13 impression input processing 14 search processing 15 storage device 20 customer terminal 21 input device 22 information processing device 23 display device 30 service center 151 product information DB 152 trust information DB 153 Custom information DB 154 Individual impression information DB 155 Individual input history DB 156 User input history DB 157 Evaluation input history DB 161 Input user input history DB

Claims (8)

[Claims] 1. An input means for inputting attribute information including at least a price and a store information of a product related to product purchase, and a voting means for handling the attribute information input to the input means as a vote for a predetermined database. The attribute information voted by the voting means is stored in the database, classified by attribute value, and the reliability of voting based on the past voting history is determined for each device or input user who has input each attribute information. Attribute information accumulating means for generating a degree, voting processing means for extracting information from the database as voting in accordance with the statistical processing in consideration of the generated reliability, and notifying the user of the voting extracted by the voting processing means. In addition to voting notification means, prices and recommended stores are classified by at least one of product category, product type, district, and store. Purchase support apparatus and outputs the broadcast as the vote. 2. The purchase support apparatus according to claim 1, wherein for each voting from the input means, the frequency of input, the difference between the input value and the displayed value of voting, the input time, and the time of voting using the voting. A purchase support apparatus comprising: a calculating unit that calculates an evaluation value related to a use value by using at least one of the presence / absence and erroneous input history, and uses the evaluation value in a statistical process for determining the vote. . 3. The purchase support apparatus according to claim 1, wherein, for each apparatus or each user who displays the vote vote extracted by the vote vote processing means, the evaluation of the user for at least each input user is held as impression information. A purchase support device comprising an individual information storage means, wherein the impression information is used in statistical processing for determining the vote. 4. The purchase support apparatus according to claim 1, wherein a predicted purchase time of the commodity is estimated based on at least one of an input cycle relating to input of attribute information accompanying the purchase of the commodity and information on a shelf life of the commodity. A purchase support apparatus comprising an estimating means, and automatically informing a user of a vote on a product to be purchased at or near a product purchase time estimated by the purchase time estimating means. 5. The purchase support apparatus according to claim 4, wherein said purchase time detecting means for detecting a purchase time at which the price of the product obtained as the vote decreases near the commodity purchase time estimated by said purchase time estimation means. A purchase support apparatus for automatically notifying a user of a vote on a product to be purchased at a time corresponding to a purchase detected by the purchase time detection means. 6. The purchase support apparatus according to claim 1, further comprising a moving cost calculating means for calculating a moving cost according to a positional relationship between a position where the user is present and a position where the product is sold, and determining the vote. A purchase support device that evaluates the information on the moving cost in the statistical processing for performing the processing. 7. The purchase support apparatus according to claim 1, further comprising inventory management means for holding information on a stock amount for each product and a storable capacity for each product on the user side, wherein the stock amount and the storable capacity are determined. A purchase support apparatus characterized by adding a determined purchaseable amount to a vote notified to a user. 8. The purchase support apparatus according to claim 1, wherein a product having an attribute indicating the purchase of the product and an attribute indicating the time of purchase among the information of the products stored in the database is purchased within a specific period. A purchase support device, comprising: a cashier managing means for notifying a user of a list of all the commodities performed as the vote.