Method, medium, and system for recommending associated products

Patent metadata

Field	Value
Publication number	US-9665894-B2
Application number	US-201213529429-A
Country	US
Kind code	B2
Filing date	Jun 21, 2012
Priority date	Dec 24, 2009
Publication date	May 30, 2017
Grant date	May 30, 2017

Abstract

A retrieval support system includes a database which stores commodity image data; a retrieval unit which acquires commodity image data having image characteristic information which is the same as or similar to image characteristic information indicating a characteristic in an image of input image data, from the database, with respect to the input image data; and a condition setting unit which sets a retrieval condition including at least one of time information relating to time, place information relating to places and occasion information relating to occasions. The retrieval unit performs retrieval of the commodity image data based on the retrieval condition set by the condition setting unit. The commodity image data acquired by the retrieval unit and information relating to a commodity different from a commodity corresponding to the commodity image data are output together.

First claim

What is claimed is:

1. A retrieval support system comprising:
a server configured to
transmit information to and receive information from at least one user terminal;
store first characteristic information in a commodity database about commodity items acquired from an image medium, each of the commodity items belonging to a category, the first characteristic information being obtained by converting elements acquired from the image medium into numeric data;
store information in a combination information database about combinations of at least one of the commodity items and a combination commodity item, the combination commodity item having a likelihood of being combined for use with the at least one of the commodity items and being included in a different category than the at least one of the commodity items;
store first attribute information in a commodity sales database relating to a group of the commodity items that are for sale;
receive input image data;
extract a contour of an image region of a desired commodity item in the input image data by performing a differential operation that calculates a difference between adjacent pixels in the input image data:
generate contour image data based on the extracted contour, the contour image data including second characteristic information, the second characteristic information being obtained by converting the elements from the input image data into the numeric data;
calculate a difference between the second characteristic information and the first characteristic information to determine a descending order of similarity between the desired commodity item and the commodity items stored in the commodity database based on an increasing difference between the second characteristic information and the first characteristic information;
select, from the commodity database, a candidate group of a predetermined number of the commodity items most similar to the desired commodity item in accordance with the descending order of similarity based on the contour image data and second attribute information about the contour image data, the contour image data representing a portion of an image input by a user, via the at least one user terminal, that corresponds to the desired commodity item, desired by the user, in the image;
retrieve, from the combination information database, the combination commodity item for each of the commodity items in the candidate group of the commodity items;
retrieve image data on the group of the commodity items that are for sale and are in the predetermined number of commodity items most similar to the combination commodity item in accordance with the descending order of similarity in one combination, selected by the user via the at least one user terminal, from the combinations of the at least one of the commodity items and the combination commodity item stored in the combination information database, according to the first attribute information about the image data of the combination commodity item of the user- selected combination and the group of the commodity items that are for sale, and output the retrieved image data as a recommended commodity;
set a retrieval condition including at least one of time information relating to time, place information relating to places, and occasion information relating to occasions,

wherein retrieval of at least one of (i) the candidate group of the commodity items, (ii) the combination commodity item for each of the commodity items in the candidate group of the commodity items, and (iii) the image data on the group of the commodity items that are for sale and are in the predetermined number of commodity items most similar to the combination commodity item in accordance with the descending order of similarity in the one combination selected by the user is additionally based on the retrieval condition.

2. The retrieval support system according to claim 1 , wherein the server is further configured to display, if a category of the desired commodity item is input by the user via the at least one user terminal, the desired commodity item in the category as a thumbnail image.

3. The retrieval support system according to claim 1 ,
wherein the predetermined number of the commodity items of the candidate group of the commodity items most similar to the desired commodity item in accordance with the descending order of similarity, which are selected based on the contour image data, are grouped according to a type, and are stored in the commodity database, and
wherein the server is further configured to extract candidate commodity items included in a same group as that of the desired commodity item as the commodity item candidate group.

4. The retrieval support system according to claim 1 ,
wherein the first attribute information is stored in the commodity database corresponding to each item of the predetermined number of the commodity items of the candidate group of the commodity items most similar to the desired commodity item in accordance with the descending order of similarity, which are selected based on the contour image data, and
wherein the server is further configured to retrieve a commodity item from the predetermined number of most similar commodity items from the commodity database in accordance with the descending order of similarity, according to the first attribute information about the commodity item of the one combination selected by the user, and extract a retrieval result as the commodity item candidate group.

5. The retrieval support system according to claim 1 ,
wherein commodity items which are in the predetermined number of commodity items most similar to the commodity items in the commodity database in accordance with the descending order of similarity are correspondingly stored in the commodity sales database, and
wherein the server is further configured to extract a commodity item which is for sale and is one of the predetermined number of commodity items most similar to the combination commodity item in accordance with the descending order of similarity in the combination selected by the user, as the recommended commodity.

6. The retrieval support system according to claim 1 ,
wherein commodity items corresponding to the commodity database are stored in the commodity sales database according to the type of each commodity item, corresponding to the first attribute information, and
wherein the server is further configured to retrieve a commodity item which is for sale and having a highest similarity in accordance with the descending order of similarity from the commodity sales database, according to the first attribute information about the combination commodity item selected by the user, and extract the retrieval result as the recommended commodity.

7. The retrieval support system according to claim 2 , the server further comprising:
a purchase history database which stores a first commodity item, purchased by the user via the at least one user terminal, corresponding to an identification signal for identifying the user,
wherein the server is configured to (i) assign an identification signal to the user who purchases a first combination commodity item, and (ii) store a first recommended commodity item and information indicating whether or not the user purchases the first commodity item in the purchase history database, corresponding to the identification signal.

8. The retrieval support system according to claim 7 ,
wherein when the server receives a user selection of the first commodity item, via the at least one user terminal, and the first recommended commodity item which is one of the predetermined number of commodity items most similar to the first commodity item in accordance with the descending order of similarity is n

Assignees

• Kumazawa Masami
• Inoue Hideya
• Yamagaki Koji
• Nikon Corp

Inventors

• Kumazawa Masami
• Inoue Hideya
• Yamagaki Koji

Classifications

• G06F16/951

  Indexing; Web crawling techniques · CPC title

• G06F16/9535

  Search customisation based on user profiles and personalisation · CPC title

• G06Q30/0601 Primary

  Electronic shopping [e-shopping] · CPC title

• G06F16/532 Primary

  Query formulation, e.g. graphical querying · CPC title