WO2015097881A1 - カテゴリ名抽出装置、カテゴリ名抽出方法及びカテゴリ名抽出プログラム - Google Patents
カテゴリ名抽出装置、カテゴリ名抽出方法及びカテゴリ名抽出プログラム Download PDFInfo
- Publication number
- WO2015097881A1 WO2015097881A1 PCT/JP2013/085166 JP2013085166W WO2015097881A1 WO 2015097881 A1 WO2015097881 A1 WO 2015097881A1 JP 2013085166 W JP2013085166 W JP 2013085166W WO 2015097881 A1 WO2015097881 A1 WO 2015097881A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- category
- categories
- category name
- product
- phrase
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/285—Clustering or classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0623—Item investigation
- G06Q30/0625—Directed, with specific intent or strategy
Definitions
- One aspect of the present invention relates to a category name extraction device, a category name extraction method, and a category name extraction program.
- the scene where the category name needs to be maintained is not limited to the electronic commerce site. For example, those that have been categorized by a person, such as classification of services and categorical search sites, need to be maintained later.
- an object of one aspect of the present invention is to reduce the time and labor required for maintenance of category names for classifying item information relating to items such as products.
- a category name extraction device includes a plurality of item information items belonging to each of a plurality of categories having a parallel relationship, and the name of the category to which each item information item belongs.
- a means for identifying a phrase common to a plurality of item information belonging to at least a plurality of different categories as a reference word and a phrase included in the item information belonging to any of the plurality of categories That extracts words / phrases that have a modified or modified relationship with the reference word / phrase and that are not names of a plurality of categories as category name candidates, and output means that outputs the category name candidates extracted by the extracting means And comprising.
- a category name extraction method is a category name extraction method in a category name extraction apparatus, and is included in a plurality of item information belonging to each of a plurality of categories in parallel relation, and each item information
- An extraction step that extracts words / phrases that are included in the word / phrase included in the information in a relationship that is modified or modified with a reference word / phrase and that is not a name of a plurality of categories, and a candidate category name extracted in the extraction step Output step.
- a category name extraction program is included in a plurality of item information items belonging to each of a plurality of categories having a parallel relationship in a computer, and the name of the category to which each item information item belongs and modification or modification A specific function that identifies, as a reference word, a phrase that is common to a plurality of item information belonging to at least a plurality of different categories, and a reference word and phrase that is included in item information belonging to one of a plurality of categories Or, an extraction function for extracting a phrase that is included in a qualified relationship and is not a name of a plurality of categories as a category name candidate, and an output function for outputting the category name candidate extracted by the extraction function are realized. .
- the phrase that is included in the item information belonging to a plurality of categories with a relationship between the name of the category to which the item information belongs and the modification or the modification, and is included in common in the item information of the plurality of different categories Is identified as a reference phrase. Then, in the item information belonging to a plurality of categories, a phrase that is included in the relationship of the reference phrase with a modification or modification is extracted and output as a category name candidate. As a result, a feature of an item belonging to the same hierarchy as a plurality of categories is expressed, and a word / phrase suitable for classifying the item is output, so that it is possible to reduce the time and effort required for maintenance such as setting a category name.
- the extracting means includes a phrase that is included in the item information belonging to a plurality of categories with a relationship with a reference word or phrase, or in a plurality of item information belonging to a plurality of categories. It is good also as excluding the phrase included more than predetermined frequency from a category name candidate.
- the word or phrase included in a plurality of item information belonging to a plurality of categories at a predetermined frequency or more. Is not a feature unique to the item, but is likely to be a phrase representing a feature common to a plurality of categories in general, and is not appropriate for a category name. According to the above aspect, it is possible to prevent a word or phrase that is not appropriate for a category name from being output as a category name candidate.
- the category name extraction apparatus may further include setting means for setting the category name candidate as a category having a parallel relationship with a plurality of categories.
- the word / phrase output as the category name candidate is set as the category, it is possible to reduce the time and labor required for setting the category.
- the setting means has a predetermined condition that the magnitude relationship between the number of searches for each category name in the plurality of categories and the number of searches for the phrase of the category name candidate in the item information belonging to the plurality of categories.
- the category name candidate may be set as a category having a parallel relationship with a plurality of categories.
- the category name candidate when a category name candidate word is set as a category, the category name candidate is set as a category when the number of items belonging to the category satisfies a predetermined condition.
- the category name candidate is set as a category when the number of items belonging to the category satisfies a predetermined condition.
- the setting means adds the category name of one category of the plurality of categories to a set of search results based on the phrase of the category name candidate in the item information belonging to the plurality of categories.
- category name candidates may be set as categories instead of one category.
- the category candidate phrases are more commonly used to represent the category than the existing category names. There is a high probability of becoming. According to the above aspect, a phrase that is more general as a phrase representing a category is set as a category, so that a category configuration suitable for searching for an item is realized.
- the setting unit includes, in the item information belonging to a plurality of categories, a predetermined description indicating a relationship between the category name of one category of the plurality of categories and the category name candidate. If included, a category name candidate may be set as a category instead of one category.
- the category name candidate is set as the category when the item information includes a description indicating the relationship between the existing category name and the category name candidate. Therefore, an appropriate phrase as the category name is set as a new category in place of the existing category.
- the plurality of categories are categories for classifying at least one of the products and services provided by the store of the electronic commerce site, and the item information is information on the products or services.
- the product is based on the number of times the product or service is purchased based on the search by the phrase of the category name candidate in the product information belonging to a plurality of categories and the product based on the search of each category name of the plurality of categories.
- the category name candidate may be set as a category having a parallel relationship with a plurality of categories.
- the plurality of categories are categories for classifying at least one of the products and services provided by the store of the electronic commerce site, and the item information is information on the products or services.
- the setting means includes a set of stores that sell products or services belonging to a plurality of categories and a set of stores that sell products or services that include the category name candidate phrase in the product information.
- a category name candidate may be set as a category.
- the category name candidate for classifying products sold by the same store as the store selling the products belonging to the existing category is set as the category, so that it has a parallel relationship with the existing category.
- An appropriate category name candidate can be set as a category as a category or a category in place of an existing category.
- the plurality of categories are categories for classifying at least one of the products and services provided by the store of the electronic commerce site, and the item information is information on the products or services.
- the product information is a product or service that includes in the product information a set of products or services sold by stores that sell products or services that belong to multiple categories, and a category name candidate phrase.
- the category name candidate may be set as a category when the set of products or services sold by the store that sells the item matches a predetermined degree or more.
- category name candidates that are classified as products sold by stores that sell products similar to those sold by stores that sell products that belong to existing categories are set as categories. Therefore, a category name candidate suitable as a category having a parallel relationship with the existing category or a category replacing the existing category can be set as the category.
- the plurality of categories are categories for classifying at least one of the products and services provided by the store of the electronic commerce site, and the item information is information on the products or services.
- the product information is the price range of products or services belonging to multiple categories and the price range of products or services that contain category name candidate phrases in the product information match more than a predetermined level.
- category name candidates may be set as categories.
- category name candidates in which products in the price range similar to the price range of products belonging to the existing category are classified are set as categories, the category or the existing category in parallel relation with the existing category is set.
- Candidate category names suitable as alternative categories can be set as categories.
- the plurality of categories are categories for classifying at least one of the products and services provided by the store of the electronic commerce site, and the item information is information on the products or services.
- the setting means may set the category name candidate as the category when the variance of the price of the product or service including the phrase of the category name candidate in the product information is equal to or less than a predetermined value.
- category name candidates that generally classify similar products when set as a category are set as categories, so that they are appropriate as categories that are parallel to existing categories or as alternatives to existing categories.
- Candidate category names can be set as categories.
- the plurality of categories are categories for classifying at least one of the products and services provided by the store of the electronic commerce site, and the item information is information on the products or services.
- the category name candidate is set as a new category by the setting means
- the product or the item including the phrase of the category name candidate in the product information or the terminal of the store selling the service Registration that changes inquiry category urging to change the category to which the service belongs to a newly set category, and changes the category of the product or the service in response to an answer from the terminal of the store in response to the transmission of the inquiry information Means may be further provided.
- the products sold by each store can be appropriately reclassified into a newly set category.
- FIG. 1 It is a figure which shows the structure of the system containing a category name extraction apparatus. It is a block diagram which shows the function structure of a category name extraction apparatus. It is a figure which shows the hardware constitutions of a category name extraction apparatus. It is a figure which shows typically the example of the category information memorize
- FIG. 1 is a diagram showing a configuration of a category name extraction system 100 including a category name extraction apparatus 1 according to the present embodiment.
- the category name extraction system 100 includes a category name extraction device 1, a user terminal T, and a store terminal D.
- the category name extraction device 1, the store terminal D, and the user terminal T are connected to each other via a network N such as the Internet.
- the store terminal D is a terminal of a store that sells products on the electronic commerce site.
- the user terminal T is a terminal held by a user who purchases a product at the electronic commerce site.
- the apparatus which comprises the shop terminal D and the user terminal T is not limited, For example, a stationary or portable personal computer may be sufficient, and portable terminals, such as a highly functional mobile phone (smart phone), a mobile phone, a personal digital assistant (PDA) But you can.
- portable terminals such as a highly functional mobile phone (smart phone), a mobile phone, a personal digital assistant (PDA) But you can.
- FIG. 2 is a block diagram showing a functional configuration of the category name extraction apparatus 1 according to the present embodiment.
- the category name extraction device 1 is a device that automatically extracts and outputs category name candidate phrases for classifying products on an electronic commerce site.
- the category name extraction apparatus 1 according to the present embodiment functionally includes a specifying unit 11 (specifying unit), an extracting unit 12 (extracting unit), an output unit 13 (output unit), and a setting unit 14. (Setting means) and a registration unit 15 (registration means).
- Each function unit of the category name extraction device 1 can access the product information storage unit 21 and the product category information storage unit 22.
- the category name extraction apparatus 1 of the present embodiment can be applied to a category name that is manually set in addition to extraction of category name candidates in a category-type search site or the like.
- FIG. 3 is a hardware configuration diagram of the category name extraction apparatus 1.
- the category name extracting device 1 is physically composed of a CPU 101, a main storage device 102 constituted by a memory such as a RAM and a ROM, an auxiliary storage device 103 constituted by a hard disk, a network card, and the like.
- the computer system includes a communication control device 104 configured, an input device 105 such as a keyboard and mouse as input devices, an output device 106 such as a display, and the like.
- Each function shown in FIG. 2 performs communication control under the control of the CPU 101 by loading predetermined computer software (category name extraction program) on the hardware such as the CPU 101 and the main storage device 102 shown in FIG. This is realized by operating the device 104, the input device 105, and the output device 106, and reading and writing data in the main storage device 102 and the auxiliary storage device 103. Data and databases necessary for processing are stored in the main storage device 102 and the auxiliary storage device 103.
- predetermined computer software category name extraction program
- the product information storage unit 21 is a storage unit that stores product information (item information) of a product (item) sold on the electronic commerce site to which the category name extraction device 1 belongs.
- the product information includes at least words related to the product.
- the product information includes a product description, a phrase indicating the product attribute, and the like.
- the product information includes information on the category to which the product belongs as an attribute.
- the product category information storage unit 22 is a storage unit that stores category information related to categories for classifying products provided by stores of the electronic commerce site.
- FIG. 4 is a diagram schematically illustrating an example of category information stored in the product category information storage unit 22.
- the category information has, for example, a tree structure (or hierarchical structure).
- the category “oil” is set below the category “skin care”.
- categories such as “jojoba”, “squalane”, and “others” are set in a lower hierarchy of the category “oil”. “Johovah”, “Squalane”, and “Others” belong to the same hierarchy and are in a parallel relationship.
- FIG. 4 is a diagram schematically illustrating an example of category information stored in the product category information storage unit 22.
- the category information has, for example, a tree structure (or hierarchical structure).
- the category “oil” is set below the category “skin care”.
- categories such as “jojoba”, “squalane”, and “others” are set in
- the category “rice” is set below the category “food”.
- categories such as “Koshihikari”, “Akitakomachi”, and “Chiba 28” are set in the lower hierarchy of the category “rice”. “Koshihikari”, “Akitakomachi”, and “Chiba 28” belong to the same hierarchy and are in a parallel relationship.
- the specifying unit 11 is a part for specifying a reference word / phrase for extracting category name candidates.
- the specifying unit 11 includes a plurality of product information (item information) belonging to each of a plurality of categories that are in a parallel relationship, and a phrase that is included in a relationship between the name of the category to which each product information belongs and a modification or a modification. Then, a phrase that is included in common in a plurality of product information belonging to at least a plurality of different categories is specified as a reference phrase.
- FIG. 5 is a diagram schematically illustrating an example of product information.
- Product information M 1 shown in FIG. 5A belongs to the category “jojoba” and includes phrases such as “jojoba oil” and “face-washing oil”.
- Product information M 2 shown in FIG. 5 (b) belongs to a category “squalane” includes phrases such as "squalane oil”, "face washing oil”.
- Product information M 3 shown in FIG. 5 (c) belong to the category "others” includes terms such "Argan oil”.
- Identification unit 11 from the product information M 1 belonging to the category "Jojoba”, acquires the phrase "Jojoba oil” including a category name "Jojoba”, from the product information M 2 belonging to the category "squalane”, categories
- the phrase “squalane oil” including the name “squalane” is acquired. That is, the specifying unit 11 acquires “jojoba oil” and “squalane oil” as words that are included in a plurality of product information belonging to each of a plurality of categories and that include a category name to which the product information belongs. Since the product information M 3 does not include a phrase including the category name “others”, the specifying unit 11 does not acquire a phrase for specifying the reference phrase from the product information M 3 .
- the specifying unit 11 extracts “oil”, which is a phrase that is included in common and modified by “jojoba” and “squalane” in the acquired phrases “jojoba oil” and “squalane oil”. Then, the specifying unit 11 specifies the extracted word “oil” as a reference word.
- the extraction unit 12 is a part that extracts, as a category name candidate, a phrase that is included in the phrase included in the product information (item information) belonging to any of a plurality of categories with a relationship of the reference phrase with the modification or modification. With reference to FIG. 5, the extraction of category name candidates will be described in detail.
- the extraction unit 12 uses the category name among the phrases “jojoba”, “face wash”, “squalane”, and “argan” that modify the reference phrase “oil” in the product information M 1 , M 2 , and M 3 . Extract “face wash” and “Argan” as category name candidates.
- the extracting unit 12 extracts the category name candidate with reference to the product information stored in the product information storage unit 21 referred to by the specifying unit 11 for specifying the reference phrase.
- the category name candidates may be extracted by referring to a product information group different from the product information group referred to by FIG.
- the extraction unit 12 may exclude words included in the plurality of product information M 1 , M 2 , and M 3 from the category name candidates that are included at a predetermined frequency or more together with the reference word / phrase. Specifically, for example, when the predetermined frequency is 2, the extraction unit 12 extracts two phrases “face wash” in the product information M 1 , M 2 , M 3 , and therefore “face wash”. Is excluded from the category name candidates. That is, based on the product information example shown in FIG. 5, the extraction unit 12 extracts a phrase such as “Argan” as a category name candidate. As a result, it is possible to prevent a word or phrase that is not appropriate for a category name from being output as a category name candidate.
- the output unit 13 is a part that outputs the category name candidates extracted by the extraction unit 12. Specifically, the output unit 13 outputs “Argan” which is a phrase of a category name candidate. Examples of output include display output for presentation to the manager of the commercial transaction management site, output to a predetermined storage means, and the like. Further, the output unit 13 may output category name candidates to the setting unit 14 for setting as a category.
- the setting unit 14 is a part for setting the category name candidate output by the output unit 13 as a category.
- the category setting will be specifically described with reference to FIGS.
- FIG. 6 is category information stored in the product category information storage unit 22 and is a diagram schematically illustrating the category information after the change in FIG. As illustrated in FIG. 6, the setting unit 14 sets a new category “Argan” that is in parallel with “jojoba” and “squalane”.
- the setting unit 14 is not an essential configuration for the present invention.
- FIG. 7 is a diagram schematically illustrating an example of product information.
- FIG. 8 is category information stored in the product category information storage unit 22 and is a diagram schematically showing the category information after the change in FIG. 4B.
- Each piece of product information shown in FIG. 7 belongs to one of the categories shown in FIG.
- Product information M 4 shown in FIG. 7 (a) belongs to the category "Koshihikari", including the phrase "Niigata Koshihikari”.
- the product information M 5 shown in FIG. 7B belongs to the category “Akitakomachi” and includes the phrase “Akitamachi from Akita”.
- the merchandise information M 7 shown in FIG. 7D belongs to the category “Chiba 28” and includes the phrase “Fusakogane (old name: Chiba 28)”.
- Specific unit 11 from the product information M 4 belonging to the category "Koshihikari”, acquires the phrase "Niigata Koshihikari” which includes a category name "Koshihikari”, from the product information M 5 belonging to the category "Akitakomachi” Acquire the phrase “Akitamachi from Akita”, which includes the category name “Akitakomachi”.
- the specifying unit 11 acquires “Niigata Koshihikari” and “Akita Akitakomachi” as terms that are included in a plurality of product information belonging to each of a plurality of categories and include the category name to which the product information belongs. To do.
- the specifying unit 11 does not acquire a phrase for specifying the reference phrase from the product information M 6 . Further, the specifying unit 11 may acquire the phrase “Fusakogane (old name: Chiba 28)” including the category name “Chiba 28”. Then, the identifying unit 11 includes “(place name) which is a phrase that modifies the phrase of the category name in the acquired phrases“ Koshihikari from Niigata ”,“ Akitamachi from Akita ”, and“ Fusakogane (former name: Chiba 28) ”. ) "Produce”. Then, since the phrase “(place name) product” is included in the product information of a plurality of categories in common, the specifying unit 11 specifies the extracted phrase “(place name) product” as the reference phrase.
- the extraction unit 12 has already set as a category name among the phrases “Koshihikari”, “Akitakomachi”, and “Fusakogane” that are qualified with the reference phrase “product of (place name)” in the product information M 4 , M 5 , M 6 “Fusakogane”, which is not a phrase that has been set, is extracted as a category name candidate.
- the output unit 13 outputs “Fusakogane” which is a phrase of the category name candidate.
- the output unit 13 outputs these category name candidates to the setting unit 14 for setting as a category.
- Category name candidates may be set as categories. Specifically, a format such as “(existing category name) (old name: (category name candidate))” is preset as a predetermined description in order to set a category name candidate phrase as a new category instead of the existing category. If the description such as “Fusakogane (old name: Chiba 28)” is extracted from the product information M 7 , the setting unit 14 replaces the category name “Chiba 28” (FIG. 4B). )), As shown in FIG. 8, the category name candidate “Fusakogane” is set as a new category. Thereby, an appropriate phrase as the category name is set as a new category in place of the existing category.
- the setting unit 14 searches the number of searches for each category name of a plurality of categories and the search for a phrase of a category name candidate in the product information belonging to a plurality of categories to which the product information referred to in the extraction of the reference word / phrase is found.
- the category name candidate may be set as a category having a parallel relation with a plurality of categories.
- the category name candidate is determined based on the minimum number of searches by the search based on phrases such as “jojoba” and “squalane”.
- the setting unit 14 sets “Argan” as a category under “Oil”, assuming that the predetermined condition is satisfied when the number of searches based on the phrase “Argan” is large. As a result, it is possible to set a category with an appropriate phrase as a new category name.
- the number of searches for searching for the category name and the category name candidate phrase in the product information can be acquired by, for example, the setting unit 14 referring to and searching the product information storage unit 21 based on each phrase.
- the setting unit 14 includes the number of times that a product is purchased based on a search based on a phrase of a category name candidate in the product information belonging to a plurality of categories to which the product information referred to in the extraction of the reference word and the category names of the plurality of categories.
- the category name candidate may be set as a category having a parallel relationship with a plurality of categories.
- a search using phrases such as “jojoba” and “squalane” representing categories provided under “oil” in the category configuration of FIG.
- the unit 14 Based on the result of, set a condition that satisfies the specified condition when the number of purchases is greater than the minimum number of purchases based on the result of a search for terms such as “Argan”
- the unit 14 sets “Argan” as a category under “Oil”.
- a category with an appropriate phrase As a result, it is possible to set a category with an appropriate phrase as a new category name.
- the number of times that a product is purchased based on a search result using a specific phrase can be obtained by referring to a database storing an access log, a product purchase history, and the like at the electronic commerce site to which the category name extraction device 1 belongs.
- the setting unit 14 adds the category name of one category among the plurality of categories to a set of search results based on the category name candidate phrases in the product information belonging to the plurality of categories to which the product information referred to in the extraction of the reference phrase belongs.
- category name candidates may be set as categories instead of one category. Specifically, for example, in the category configuration of FIG.
- the setting unit 14 is a set of stores selling products belonging to a plurality of categories to which the product information referred to in the extraction of the reference word / phrase is extracted, and stores selling products including the word / phrase of category name candidates in the product information
- the category name candidate may be set as a category when the set matches a predetermined degree or more. Specifically, for example, at the electronic commerce site to which the category name extraction device 1 belongs, the products belonging to the categories “Koshihikari” and “Akitakomachi” provided under “rice” in the category configuration of FIG.
- a list of stores that sell specific products is, for example, a database (for example, product information storage unit 21) that stores product information sold by each store in the electronic commerce site to which the category name extraction device 1 belongs. Is obtained by referring to.
- the setting unit 14 includes, in the product information, a set of products sold by a store that sells products that belong to a plurality of categories to which the product information referred to in the extraction of the reference word / phrase is included, and a category name candidate word / phrase.
- the category name candidate may be set as a category when the set of products sold by the store that sells the item matches a predetermined degree or more. Specifically, for example, at the electronic commerce site to which the category name extraction device 1 belongs, the products belonging to the categories “Koshihikari” and “Akitakomachi” provided under “rice” in the category configuration of FIG.
- the setting unit 14 sets “Fusakogane” as a category under “rice”.
- a list of products sold by a store selling a specific product is, for example, a database (for example, a database storing product information sold by each store on the electronic commerce site to which the category name extraction device 1 belongs). It is obtained by referring to the merchandise information storage unit 21). Similarity between product sets can be calculated by a well-known analysis technique.
- the setting unit 14 matches the price range of the product belonging to the plurality of categories to which the product information referred to in the extraction of the reference word and the price range of the product including the category name candidate word / phrase in the product information more than a predetermined degree.
- the category name candidate may be set as a category. Specifically, for example, in the electronic commerce site to which the category name extraction device 1 belongs, in the category configuration of FIG.
- the setting unit 14 sets “Fusakogane” as a category under “rice”. Set as.
- For the price range of a specific product for example, by referring to a database (for example, product information storage unit 21) that stores product information sold by each store in the electronic commerce site to which the category name extraction device 1 belongs. can get.
- the degree of matching of price ranges can be calculated by a well-known analysis technique.
- the setting unit 14 may set the category name candidate as the category when the variance of the price of the product including the phrase of the category name candidate in the product information is equal to or less than a predetermined value. Specifically, for example, in the electronic commerce site to which the category name extraction apparatus 1 belongs, the setting unit 14 when the variance of the price of the product including the category name candidate “Fusakogane” in the product information is equal to or less than a predetermined value. Sets “Fusakogane” as a category under “rice”.
- the distribution of the price of a specific product is well-known by referring to a database (for example, product information storage unit 21) storing product information sold by each store in the electronic commerce site to which the category name extraction device 1 belongs. It can be calculated by using the statistical method.
- the registration unit 15 stores stores that sell products that include the category name candidate words in the product information.
- inquiry information that prompts the terminal D to change the category to which the product belongs to a newly set category, and changing the category of the product in response to an answer from the store terminal D in response to the transmission of the inquiry information is there.
- the registration unit 15 extracts product information including “Argan” from the product information storage unit 21. Then, the registration unit 15 transmits inquiry information that prompts the user to change the category to which the product belongs to a newly set category to the store terminal D of the store that sells the product with the extracted product information.
- FIG. 9 is a diagram illustrating an example of an inquiry information display screen. As shown in FIG. 9, the inquiry information includes a product list of product information including “Argan”, a message for prompting a category change, an operation unit for accepting the category change, and the like.
- the shop terminal D checks the check box of the product whose category is to be changed in the display screen example shown in FIG. 9 and operates the button displayed as “Re-register”, the product of the checked product is displayed.
- An answer to the effect of changing the category is returned from the store terminal D to the registration unit 15.
- the registration part 15 changes the category of the goods to which the check box was attached
- FIG. The change of the category is realized by rewriting the attribute of the product information of the product stored in the product information storage unit 21. As a result, the products sold by each store can be appropriately reclassified into the newly set category.
- FIG. 10 is a flowchart showing an example of processing contents of the category name extraction method in the category name extraction apparatus 1 shown in FIG.
- the specifying unit 11 acquires a plurality of pieces of product information belonging to each of a plurality of categories having a parallel relationship (S1). Next, the specifying unit 11 acquires a phrase that is included in the phrase included in the acquired merchandise information together with a category name of the category to which the merchandise information belongs and a modification or modification (S2).
- the specifying unit 11 specifies the phrase that is acquired in step S2 and is commonly included in the product information belonging to each of the plurality of categories as the reference phrase (S3).
- the extraction unit 12 uses, as a category name candidate, a word that is included in the product information belonging to one of a plurality of categories together with a reference word or phrase that has a modification or modification relationship and is not set as a category name. Extract (S4). Then, the output unit 13 outputs the category name candidates extracted by the extraction unit 12 (S5).
- the category name extraction program 1p includes a main module 10m, a specific module 11m, an extraction module 12m, an output module 13m, a setting module 14m, and a registration module 15m.
- the main module 10m is a part that comprehensively controls the category name extraction process.
- the functions realized by executing the identification module 11m, the extraction module 12m, the output module 13m, the setting module 14m, and the registration module 15m are respectively the identification unit 11, the extraction unit 12, and the category name extraction device 1 shown in FIG.
- the functions of the output unit 13, the setting unit 14, and the registration unit 15 are the same.
- the category name extraction program 1p is provided by a storage medium 1d such as a CD-ROM or a DVD-ROM or a semiconductor memory, for example.
- the category name extraction program 1p may be provided via a communication network as a computer data signal superimposed on a carrier wave.
- the category name to which the product information belongs to the word / phrase included in the product information belonging to a plurality of categories is qualified or covered. Phrases that are included in a modification relationship and that are commonly included in product information of a plurality of different categories are specified as reference phrases. Then, in the product information belonging to a plurality of categories, phrases included in the relationship between the reference phrase and the modification or modification are extracted and output as category name candidates. As a result, the characteristics of the products belonging to the same hierarchy as the plurality of categories are expressed, and words suitable for classifying the products are output, so that it is possible to reduce the time and effort required for maintenance such as setting category names.
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Economics (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Human Resources & Organizations (AREA)
- General Engineering & Computer Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Game Theory and Decision Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Description
Claims (14)
- 並列関係にある複数のカテゴリのそれぞれに属する複数のアイテム情報に含まれており、各アイテム情報が属するカテゴリの名称と修飾又は被修飾の関係をもっており、少なくとも複数の異なるカテゴリに属する複数のアイテム情報に共通する語句を、基準語句として特定する特定手段と、
前記複数のカテゴリのいずれかに属するアイテム情報に含まれる語句に前記基準語句と修飾又は被修飾の関係をもって含まれる語句であって前記複数のカテゴリの名称ではない語句をカテゴリ名候補として抽出する抽出手段と、
前記抽出手段により抽出されたカテゴリ名候補を出力する出力手段と、
を備えるカテゴリ名抽出装置。 - 前記抽出手段は、前記複数のカテゴリに属するアイテム情報に前記基準語句と修飾又は被修飾の関係をもって含まれる語句であって、前記複数のカテゴリに属する複数のアイテム情報において所定の頻度以上で含まれる語句をカテゴリ名候補から除外する、
請求項1に記載のカテゴリ名抽出装置。 - 前記カテゴリ名候補を前記複数のカテゴリと並列関係となるカテゴリとして設定する設定手段を更に備える、
請求項1または2に記載のカテゴリ名抽出装置。 - 前記設定手段は、前記複数のカテゴリに属するアイテム情報における、前記複数のカテゴリの各カテゴリ名の検索数と前記カテゴリ名候補の語句の検索数との大小関係が所定条件を満たす場合に、前記カテゴリ名候補を前記複数のカテゴリと並列関係となるカテゴリとして設定する、
請求項3に記載のカテゴリ名抽出装置。 - 前記設定手段は、前記複数のカテゴリに属するアイテム情報における、前記カテゴリ名候補の語句による検索結果の集合に、前記複数のカテゴリのうちの一のカテゴリのカテゴリ名の語句による検索結果の集合が所定程度以上含まれる場合に、前記一のカテゴリに代えて前記カテゴリ名候補をカテゴリとして設定する、
請求項3または4に記載のカテゴリ名抽出装置。 - 前記設定手段は、前記複数のカテゴリに属するアイテム情報に、前記複数のカテゴリのうちの一のカテゴリのカテゴリ名と前記カテゴリ名候補との関係を示す所定の記載が含まれる場合に、前記一のカテゴリに代えて前記カテゴリ名候補をカテゴリとして設定する、
請求項3~5のいずれか1項に記載のカテゴリ名抽出装置。 - 前記複数のカテゴリは、電子商取引サイトの店舗が提供する商品及び役務の少なくとも1つを分類するためのカテゴリであり、
前記アイテム情報は、商品または役務に関する情報である商品情報であり、
前記設定手段は、前記複数のカテゴリに属する商品情報における、前記カテゴリ名候補の語句による検索に基づき商品または役務が購入された回数と前記複数のカテゴリの各カテゴリ名の検索に基づき商品または役務が購入された回数との大小関係が所定条件を満たす場合に、前記カテゴリ名候補を前記複数のカテゴリと並列関係となるカテゴリとして設定する、
請求項3~6のいずれか1項に記載のカテゴリ名抽出装置。 - 前記複数のカテゴリは、電子商取引サイトの店舗が提供する商品及び役務の少なくとも1つを分類するためのカテゴリであり、
前記アイテム情報は、商品または役務に関する情報である商品情報であり、
前記設定手段は、前記複数のカテゴリに属する商品または役務を販売している店舗の集合と、前記カテゴリ名候補の語句を商品情報に含む商品または役務を販売している店舗の集合と、が所定の程度以上一致する場合に、前記カテゴリ名候補をカテゴリとして設定する、
請求項3~7のいずれか1項に記載のカテゴリ名抽出装置。 - 前記複数のカテゴリは、電子商取引サイトの店舗が提供する商品及び役務の少なくとも1つを分類するためのカテゴリであり、
前記アイテム情報は、商品または役務に関する情報である商品情報であり、
前記設定手段は、前記複数のカテゴリに属する商品または役務を販売している店舗により販売されている商品または役務の集合と、前記カテゴリ名候補の語句を商品情報に含む商品または役務を販売している店舗により販売されている商品または役務の集合と、が所定の程度以上一致する場合に、前記カテゴリ名候補をカテゴリとして設定する、
請求項3~8のいずれか1項に記載のカテゴリ名抽出装置。 - 前記複数のカテゴリは、電子商取引サイトの店舗が提供する商品及び役務の少なくとも1つを分類するためのカテゴリであり、
前記アイテム情報は、商品または役務に関する情報である商品情報であり、
前記設定手段は、前記複数のカテゴリに属する商品または役務の価格帯と、前記カテゴリ名候補の語句を商品情報に含む商品または役務の価格帯と、が所定の程度以上一致する場合に、前記カテゴリ名候補をカテゴリとして設定する、
請求項3~9のいずれか1項に記載のカテゴリ名抽出装置。 - 前記複数のカテゴリは、電子商取引サイトの店舗が提供する商品及び役務の少なくとも1つを分類するためのカテゴリであり、
前記アイテム情報は、商品または役務に関する情報である商品情報であり、
前記設定手段は、前記カテゴリ名候補の語句を商品情報に含む商品または役務の価格の分散が所定値以下である場合に、前記カテゴリ名候補をカテゴリとして設定する、
請求項3~10のいずれか1項に記載のカテゴリ名抽出装置。 - 前記複数のカテゴリは、電子商取引サイトの店舗が提供する商品及び役務の少なくとも1つを分類するためのカテゴリであり、
前記アイテム情報は、商品または役務に関する情報である商品情報であり、
前記設定手段により前記カテゴリ名候補が新たなカテゴリとして設定される場合に、前記カテゴリ名候補の語句を商品情報に含む商品または役務を販売している店舗の端末に、該商品または該役務が属するカテゴリを新たに設定されるカテゴリに変更することを促す問合せ情報を送信し、該問合せ情報の送信に対する前記店舗の端末からの回答に応じて該商品または該役務のカテゴリを変更する登録手段を更に備える、
請求項3~11のいずれか1項に記載のカテゴリ名抽出装置。 - カテゴリ名抽出装置におけるカテゴリ名抽出方法であって、
並列関係にある複数のカテゴリのそれぞれに属する複数のアイテム情報に含まれており、各アイテム情報が属するカテゴリの名称と修飾又は被修飾の関係をもっており、少なくとも複数の異なるカテゴリに属する複数のアイテム情報に共通する語句を、基準語句として特定する特定ステップと、
前記複数のカテゴリのいずれかに属するアイテム情報に含まれる語句に前記基準語句と修飾又は被修飾の関係をもって含まれる語句であって前記複数のカテゴリの名称ではない語句をカテゴリ名候補として抽出する抽出ステップと、
前記抽出ステップにおいて抽出されたカテゴリ名候補を出力する出力ステップと、
を有するカテゴリ名抽出方法。 - コンピュータに、
並列関係にある複数のカテゴリのそれぞれに属する複数のアイテム情報に含まれており、各アイテム情報が属するカテゴリの名称と修飾又は被修飾の関係をもっており、少なくとも複数の異なるカテゴリに属する複数のアイテム情報に共通する語句を、基準語句として特定する特定機能と、
前記複数のカテゴリのいずれかに属するアイテム情報に含まれる語句に前記基準語句と修飾又は被修飾の関係をもって含まれる語句であって前記複数のカテゴリの名称ではない語句をカテゴリ名候補として抽出する抽出機能と、
前記抽出機能により抽出されたカテゴリ名候補を出力する出力機能と、
を実現させるカテゴリ名抽出プログラム。
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2013/085166 WO2015097881A1 (ja) | 2013-12-27 | 2013-12-27 | カテゴリ名抽出装置、カテゴリ名抽出方法及びカテゴリ名抽出プログラム |
EP13900381.8A EP3089096A4 (en) | 2013-12-27 | 2013-12-27 | Category name extraction device, category name extraction method and category name extraction program |
JP2014510593A JP5530047B1 (ja) | 2013-12-27 | 2013-12-27 | カテゴリ名抽出装置、カテゴリ名抽出方法及びカテゴリ名抽出プログラム |
US14/758,318 US10621208B2 (en) | 2013-12-27 | 2013-12-27 | Category name extraction device, category name extraction method, and category name extraction program |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2013/085166 WO2015097881A1 (ja) | 2013-12-27 | 2013-12-27 | カテゴリ名抽出装置、カテゴリ名抽出方法及びカテゴリ名抽出プログラム |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2015097881A1 true WO2015097881A1 (ja) | 2015-07-02 |
Family
ID=51175834
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2013/085166 WO2015097881A1 (ja) | 2013-12-27 | 2013-12-27 | カテゴリ名抽出装置、カテゴリ名抽出方法及びカテゴリ名抽出プログラム |
Country Status (4)
Country | Link |
---|---|
US (1) | US10621208B2 (ja) |
EP (1) | EP3089096A4 (ja) |
JP (1) | JP5530047B1 (ja) |
WO (1) | WO2015097881A1 (ja) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2012116208A2 (en) * | 2011-02-23 | 2012-08-30 | New York University | Apparatus, method, and computer-accessible medium for explaining classifications of documents |
US20170270577A1 (en) * | 2016-03-15 | 2017-09-21 | Ebay Inc. | Catalogue management |
US11588949B1 (en) * | 2021-08-16 | 2023-02-21 | Toshiba Tec Kabushiki Kaisha | Image forming apparatus and conveyance control method |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2005107937A (ja) * | 2003-09-30 | 2005-04-21 | Fujitsu Ltd | 商品情報管理用プログラム,それを格納したコンピュータ可読媒体,及び、それが取り扱う商品分類マスタデータベースのデータ構造 |
JP2007505422A (ja) | 2003-06-13 | 2007-03-08 | シーエヌイーティ ネットワークス インコーポレイテッド | 製品情報を格納するためのカタログ分類装置、該カタログ分類装置を使用するシステム及び方法 |
JP2008097520A (ja) * | 2006-10-16 | 2008-04-24 | Denso Corp | 検索装置 |
JP5308593B2 (ja) * | 2011-07-25 | 2013-10-09 | 楽天株式会社 | ジャンル生成装置 |
Family Cites Families (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6366910B1 (en) * | 1998-12-07 | 2002-04-02 | Amazon.Com, Inc. | Method and system for generation of hierarchical search results |
US8019659B2 (en) | 2003-05-02 | 2011-09-13 | Cbs Interactive Inc. | Catalog taxonomy for storing product information and system and method using same |
CN1629837A (zh) * | 2003-12-17 | 2005-06-22 | 国际商业机器公司 | 电子文档的处理、浏览及分类查询的方法、装置及其系统 |
US7885859B2 (en) * | 2006-03-10 | 2011-02-08 | Yahoo! Inc. | Assigning into one set of categories information that has been assigned to other sets of categories |
KR101049889B1 (ko) * | 2007-10-22 | 2011-07-19 | 주식회사 이베이지마켓 | 검색을 통한 행태분석에 기반한 키워드 그룹에 대하여광고를 수주하고 타겟 광고하는 웹 사이트 운영 방법 및온라인 시스템 |
US8805823B2 (en) * | 2009-04-14 | 2014-08-12 | Sri International | Content processing systems and methods |
US8489523B2 (en) * | 2010-03-31 | 2013-07-16 | Alcatel Lucent | Categorization automation based on category ontology |
CN102541862B (zh) * | 2010-12-14 | 2014-05-07 | 阿里巴巴集团控股有限公司 | 跨网站的信息显示方法及系统 |
US9171088B2 (en) * | 2011-04-06 | 2015-10-27 | Google Inc. | Mining for product classification structures for internet-based product searching |
US9201967B1 (en) * | 2012-05-10 | 2015-12-01 | Amazon Technologies, Inc. | Rule based product classification |
-
2013
- 2013-12-27 WO PCT/JP2013/085166 patent/WO2015097881A1/ja active Application Filing
- 2013-12-27 EP EP13900381.8A patent/EP3089096A4/en not_active Withdrawn
- 2013-12-27 JP JP2014510593A patent/JP5530047B1/ja active Active
- 2013-12-27 US US14/758,318 patent/US10621208B2/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2007505422A (ja) | 2003-06-13 | 2007-03-08 | シーエヌイーティ ネットワークス インコーポレイテッド | 製品情報を格納するためのカタログ分類装置、該カタログ分類装置を使用するシステム及び方法 |
JP2005107937A (ja) * | 2003-09-30 | 2005-04-21 | Fujitsu Ltd | 商品情報管理用プログラム,それを格納したコンピュータ可読媒体,及び、それが取り扱う商品分類マスタデータベースのデータ構造 |
JP2008097520A (ja) * | 2006-10-16 | 2008-04-24 | Denso Corp | 検索装置 |
JP5308593B2 (ja) * | 2011-07-25 | 2013-10-09 | 楽天株式会社 | ジャンル生成装置 |
Non-Patent Citations (1)
Title |
---|
See also references of EP3089096A4 |
Also Published As
Publication number | Publication date |
---|---|
JPWO2015097881A1 (ja) | 2017-03-23 |
EP3089096A1 (en) | 2016-11-02 |
JP5530047B1 (ja) | 2014-06-25 |
US10621208B2 (en) | 2020-04-14 |
EP3089096A4 (en) | 2017-05-10 |
US20150347564A1 (en) | 2015-12-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108346075B (zh) | 信息推荐方法和装置 | |
US20150074114A1 (en) | Tag management device, tag management method, tag management program, and computer-readable recording medium for storing said program | |
US20170097940A1 (en) | Analytical Search Engine | |
US20160335693A1 (en) | Information providing device, information providing method, program and non-transitory recording medium | |
US20230089850A1 (en) | Real-time product environmental impact scoring | |
KR20180052489A (ko) | 사용자 경험분석 및 환경요인에 기초한 크로스보더 전자상거래 상품 추천 방법 | |
AU2018211215A1 (en) | Method and system for modifying a webpage | |
CN116308684B (zh) | 一种网购平台店铺信息推送方法及系统 | |
KR101509393B1 (ko) | 상품 정보 제공 장치 및 그 방법 | |
JP5530047B1 (ja) | カテゴリ名抽出装置、カテゴリ名抽出方法及びカテゴリ名抽出プログラム | |
Zhao et al. | Anatomy of a web-scale resale market: a data mining approach | |
JP5567749B2 (ja) | 辞書生成装置、辞書生成方法、辞書生成プログラム、及びそのプログラムを記憶するコンピュータ読取可能な記録媒体 | |
JP5265414B2 (ja) | ネットショッピング管理装置 | |
JP5568195B1 (ja) | 検索システム、検索条件設定装置、検索条件設定装置の制御方法、プログラム、及び情報記憶媒体 | |
CN112488854A (zh) | 服务经理个性化推荐方法和相关设备 | |
JP5670490B2 (ja) | カテゴリ判定装置、検索装置、カテゴリ判定方法、カテゴリ判定プログラム、及びそのプログラムを記憶するコンピュータ読取可能な記録媒体 | |
KR20180092053A (ko) | 지능형 쇼핑 관리시스템 | |
US11308941B2 (en) | Natural language processing apparatus and program | |
Arnold et al. | Semi-automatic identification of counterfeit offers in online shopping platforms | |
KR102415016B1 (ko) | 온라인 매크로 시스템의 동작 방법 | |
JP7260294B2 (ja) | 情報処理装置、情報処理方法及び情報処理プログラム | |
JP2010122756A (ja) | Edi統合処理システム、edi統合処理方法、およびedi統合処理プログラム | |
TW202418187A (zh) | 電商商品搜尋系統及其依照情境搜尋電商商品之方法 | |
CN112907311A (zh) | 物品的识别方法及装置、计算机存储介质、电子设备 | |
CN117556474A (zh) | 数据处理方法、装置、计算机设备和存储介质 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
ENP | Entry into the national phase |
Ref document number: 2014510593 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 14758318 Country of ref document: US |
|
REEP | Request for entry into the european phase |
Ref document number: 2013900381 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2013900381 Country of ref document: EP |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 13900381 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |