WO2011134950A1 - Processing input data comprising references to products or services - Google Patents

Processing input data comprising references to products or services Download PDF

Info

Publication number
WO2011134950A1
WO2011134950A1 PCT/EP2011/056555 EP2011056555W WO2011134950A1 WO 2011134950 A1 WO2011134950 A1 WO 2011134950A1 EP 2011056555 W EP2011056555 W EP 2011056555W WO 2011134950 A1 WO2011134950 A1 WO 2011134950A1
Authority
WO
WIPO (PCT)
Prior art keywords
input data
item
matching result
information
item identifier
Prior art date
Application number
PCT/EP2011/056555
Other languages
French (fr)
Inventor
Michel Cassius
Rodrigo Dauster
Romain Eude
Original Assignee
Gekko.Com Limited
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Gekko.Com Limited filed Critical Gekko.Com Limited
Publication of WO2011134950A1 publication Critical patent/WO2011134950A1/en

Links

Classifications

    • 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]
    • G06Q30/0603—Catalogue ordering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/3331—Query processing
    • G06F16/334—Query execution
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/279—Recognition of textual entities
    • G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
    • G06F40/295—Named entity recognition

Definitions

  • Processing input data comprising references to products or services
  • the present invention relates to online searching for information about products or services. More specifically, the invention relates to processing input data comprising references to products or services.
  • a computer- implemented method in a server for processing input data.
  • the input data comprises a reference to a product or a service.
  • the server comprising a database.
  • the database comprising a reference to a product or a service.
  • the method comprises receiving the input data from an end-user device.
  • the method further comprises searching through the input data for a first part of the input data that matches with an item identifier or a variant of the item identifier to obtain a first matching result.
  • the method further comprises, if the first part of the input data matched with two or more item identifiers or a variant of two or more item identifiers, searching through the input data for a second part of the input data that matches with a contextual reference to obtain a second matching result and selecting one item identifier as the first matching result based on the second matching result.
  • the method further comprises processing based on the first matching result the first part of the input data as the reference to the product or the service.
  • the input data provided by an end-user through his end-user device can be analysed by the server for known products and/or services. If a reference to a product or a service is found in the input data, i.e. when the first part is found, then the reference in the input data is processed to help the end-user to find relevant information about the product or service .
  • the embodiment of claim 2 advantageously enables a better matching of the first part with an item identifier.
  • the embodiment of claim 3 advantageously enables a URL, i.e. an uniform resource locator such as a web site address, to be used as input data.
  • a URL i.e. an uniform resource locator such as a web site address
  • the embodiment of claim 4 advantageously enables spoken input to be used as input data.
  • the embodiment of claim 5 advantageously enables a text to be used as input data.
  • the embodiment of claim 6 advantageously enables a digital image or bar code to be used as input data.
  • the embodiment of claim 7 advantageously enables web pages to be processed and modified to help the end-user to find relevant information about the product or service using the web page .
  • the embodiment of claim 8 advantageously enables limited number of references to be flagged on a web page to avoid overcrowding the webpage with visual cues.
  • the embodiment of claim 9 advantageously enables the generation of an online overview of information about product and services.
  • the embodiment of claim 10 advantageously enables the end-user to receive information directly.
  • the embodiment of claim 11 advantageously enables various products or services to be compared.
  • the embodiment of claim 12 advantageously the acquired product and service information to be consulted by the end-user at any time and on any device.
  • a computer program element is proposed, which, when being executed by a.
  • processor is adapted to carry out one or more of the steps of the method above .
  • a server is proposed to carry our one or more of the steps of the method above .
  • Fig.l illustrates a flow diagram of an exemplary embodiment of the invention
  • Fig.2 illustrates a network architecture of an exemplary embodiment of the invention
  • Fig 3 illustrates a flow diagram of a process for the consumer to collect and collate products and/or services of an exemplary embodiment of the invention.
  • the invention enables information about products and/or services to be found more easily and enhanced. In one embodiment it becomes possible to collate the information on various related products and/or services to enable the user to compare these various products or services, e.g. to make a purchase decision.
  • the invention enables publishers to more easily monetize their content and retain their users whilst providing users with a more convenient and more intuitive buying experience.
  • the input data can be a web page retrieved by the user using a web browser on an end-user device 20, which is analysed by embedded code in the web page or a browser plug-in in connection with a server 10.
  • the input data can be a reference of a particular product or service provided by the end-user device 20 to the server 10 in the form of a text, URL, bar code, digital image or digitized voice.
  • the user can provide the reference though a variety of communication channels 20 which may include, but are not limited to, email, text messaging, message stream, instant messaging, or an electronic interchange via an application programming interface (API) and using a variety of end-user devices which may include, but are not limited to, PC's 21, smartphones 22, tablet PC's 23, notebooks 24, set-top boxed 25 and online televisions 26.
  • a variety of communication channels 20 which may include, but are not limited to, email, text messaging, message stream, instant messaging, or an electronic interchange via an application programming interface (API)
  • end-user devices which may include, but are not limited to, PC's 21, smartphones 22, tablet PC's 23, notebooks 24, set-top boxed 25 and online televisions 26.
  • the link to the URL is followed to get the referenced web page as input data.
  • URLs in web pages may be followed as well, in which case the
  • referenced web page may be added to the input data.
  • the input data is searched for references matching an item identifier.
  • the item identifier is a name, product id, SKU number, model number or any other unique identifier of the product or service.
  • the input data is parsed and matched against a
  • synchronised database 11 of the server 10 containing product and service item identifiers.
  • the input data is a text or web page, then its content is e.g. parsed and matched against a synchronised database of product and service item identifiers.
  • the input data is e.g. a bar code, then the bar code is matched against a synchronised database of product or service bar codes and item identifiers.
  • the input data is e.g. an image or an image on a web page, then the image is matched against a synchronised database of product or service images and item identifiers.
  • the message format is e.g. digitized voice, then the speech is converted to text and matched a synchronised database of product or service item identifiers.
  • the database typically contains variations of item identifiers to be found in the input data.
  • variations of item identifiers may all be matched to "The Sanderson Hotel”.
  • image matching different images of the same product or service or any known advanced image recognition technology may be used to match the image to an item identifier.
  • the database may contain generic words, such as "The”, “Hotel”, “Spa”, “Resort” or “Inn” and specific words, such as "Sanderson".
  • the generic words can be detected in the input data and combined with a specific word to find the correct item identifier. E.g. a matching of the generic word “The” and/or “Hotel” near the specific word “Sanderson” results in the item identifier "The Sanderson Hotel”.
  • references to destinations in the input data may be used to increase the confidence that the correct item identifier is found.
  • the references to destination are stored in the database in relation to the item identifiers.
  • the references contain e.g. countries, cities and/or street names. If e.g. in the input data the text "Ritz” is found and a reference in the form of the text "London”, then the item identifier is set to the Ritz hotel in London as opposed to e.g. the Ritz hotel in New York.
  • the reference recognition algorithm identifies ambiguous and unambiguous phrases that are direct or indirect references to products or services on any input data using semantic rules that consider various permutations of words and symbols, and searches for a context within which they are mentioned so as to improve recognition. For example the phrase “When I visit London I like staying in the Mandarin” leads to the identification of the “Mandarin Oriental Hyde Park”, the phrase “It is rumoured that Nikon is going to announce a new camera to replace the DX40" leads to the identification of the "Nikon DX-40" as opposed to e.g. the "Heathkit DX40" or the "Magnum DX40".
  • the input data can be enhanced by adding information to the reference.
  • Fig.l shows a flow diagram of an exemplary embodiment of the invention.
  • the input data is received from an end-user device in the server.
  • the server searches through the input data for a first part of the input data that matches with an item identifier or a variant of the item
  • step 3 it is determined if the first part of the input data matched with two or more item identifiers or a variant of two or more item identifiers. If this is the case, then in step 4 the server searches through the input data for a second part of the input data that matches with a contextual reference to obtain a second matching result and selecting one item identifier as the first matching result based on the second matching result. In step 5, based on the first matching result, the first part of the input data is processed as the reference to the product or the service .
  • the input data i an online document such as a web page wherein one or more references to products or services are to be flagged. Flagging may be done by emphasizing the reference by changing its colour and/or font face or by converting into a hyperlink to other web pages, possibly on other web sites.
  • the online document can be user-generated or machine-generated.
  • the references can be text- based or image-based.
  • the reference is recognised by comparing the text and images in the online document against a synchronised database of past and present products and services, typically within a given market segment.
  • the identification of the online document to be processed can be chosen either by the publisher of the document by adding program instructions in the form of embedded code to the document or by the user though the use of a browser plug-in.
  • embedded code are JavaScript code, Flash code or any other suitable code.
  • An example of a browser plug-in is ibid.
  • the reference is flagged e.g. by adding a
  • the hyperlink is e.g. an affiliate link.
  • the address of the link may vary
  • the web page, or part of the web page, is typically scraped prior to analysing by removing portions irrelevant to the content itself, such as header parts and tags.
  • references should be flagged to the user and how to direct their actions based on dynamic business rules which include one or more of, but are not limited to: how many references are found; how many different references exist on the page; how many instances of a given reference appear on the page; which part(s) of the page the reference appears in; and the context of the request (e.g. time of day and location of user) .
  • a flag may be added allowing the user to access more information about the product only for two references. For example, for a given publisher, only flag references in the main body of the page may be added, i.e. not on comments or items appearing on the side columns.
  • the server communicates to the user's browser the location of the targeted references on the web page the user is visiting and what action to take if the user interacts with it such as displaying on the web page additional information about the reference, providing transaction or ecommerce options, or sending the user, via a hyperlink, to another web page with related information.
  • Visual cues can be added to the targeted references to call the attention of the user.
  • visual cues are hyperlinks, colour highlight, changing text display colour or style and adding an. icon.
  • a preset action is performed, such as following the link or displaying additional information in a pop-up.
  • the item identifier found in the input data is used to enable users to collect references of products or services of interest.
  • consumers collect references to products or services of interest. Once collected, information about these products or services is enhanced, standardised and normalised. Tools are offered for the consumer to collate the products and services they collected and present the information.
  • shortlist identifiers are captured.
  • the consumer indicates one or more products or services of interest for collecting and collating.
  • the consumer provides the shortlist items in input data using any known message format and communication channel.
  • the message formats may include, but are not limited to text, URLs, bar codes, images or voice.
  • the communication channels may include, but are not limited to email, text message, instant message, electronic interchange via an application programming interface (API) .
  • API application programming interface
  • the consumer can indicate the shortlist items directly or indirectly.
  • Examples of a direct shortlist item are an email with the content "There are two hotels in London I am trying to decide between: the Sanderson and st martins [sic] lane” and a tweet with the content "@michel do you recommend the Sanderson”.
  • Examples of indirect shortlist items are an URL or a public social media message stream, which can be parsed for references to items consumers may want to shortlist.
  • a second step 102 the item identifiers are identified in the input data.
  • the input data is analysed to determine the relevant item identifier. If the message format is e.g. an URL, then the link is followed to get the item
  • the message format is e.g. a non-linked text
  • the input data is parsed and matches against a synchronised database of product or service item identifiers.
  • the message format is e.g. an image
  • the input data is matched against a synchronised database of product or service images and item identifiers.
  • the message format is e.g. voice
  • the input data is converted to text and matched against a synchronised, database of product or service item identifiers.
  • Each shortlist item may contain more than one item identifier.
  • each item identifier has an associated accuracy probability score of how well it matches the shortlist item.
  • the accuracy probability score it is determined whether to select the item identifier with the highest accuracy probability score or whether to request the consumer to clarify their shortlist item.
  • a third step 103 the shortlist item information is enhanced. Given the item identifier, a multitude of information may be aggregated about the shortlist items.
  • Sources of information may include, but are not limited to, information provided by the provider of the product or service, a synchronised database of past and present products and services, information provided by other vendors of the product or service and/or consumer reviews.
  • the information may include hyperlinks possibly including affiliate links to other sites where the consumer can learn more about and purchase or consumption of the shortlist item, prices, features, functionality, images, user reviews, and/or any other information.
  • a response from the server is provided to the user at this stage.
  • the server may response with the tweet: "@rodrigo you can find more about the Sanderson here www. ko . com/j 8HgT" .
  • a fourth step 104 the item information is standardised, normalised and organised.
  • Information on the products or services is standardised, normalised and organised to provide like ⁇ for ⁇ like comparison possibilities between different products or services.
  • a fifth step 105 the shortlist items and all associated information are stored.
  • the associated information includes the item identifier, the item information and any additional content contributed by the consumer and possibly by other consumers .
  • the additional content contributed by the consumer may include comments and annotations, indication of preferences through some form of polling and tracked usage behaviour such as views, duration of view and what content the consumers clicked on, and other information.
  • a sixth step 106 the shortlist items and all associated information are presented.
  • the consumer can view and optionally modify the shortlist items and contribute additional content through a multitude of portals or devices that include, but are not limited to, computers, computing devices mobile devices and televisions.
  • portals or devices include, but are not limited to, computers, computing devices mobile devices and televisions.
  • the organisation and presentation of the item information is typically adjusted to the consumer's device capabilities.
  • the consumer can optionally invite other individuals, each with distinct access rights, to view and/or modify the shortlist items and contribute additional content to create a shared collaborative document.
  • ft server saves the chosen products and services in a list that belongs to the consumer and has customisable access rights which include but are not limited to: the consumer can choose who can see the list; the list can be private (viewable only by chosen consumers) or public (viewed by many) ; the consumer can choose who, other than themselves, can add items to the list; the consumer can choose who, other than themselves, can delete items from the list; the consumer can change any of these access rights at any time without loss of any information; others .
  • the invention can be used for information about any product or service.
  • the invention can be used for the hotels market, the restaurant market, or other specific application uses .
  • One embodiment of the invention may be implemented as a program product for use with a computer system.
  • the program (s) of the program product define functions of the embodiments
  • Illustrative computer-readable storage media include, but are not limited to: (i) non-writable storage media (e.g., read-only memory devices within a computer such as CD-ROM disks readable by a CD-ROM drive, ROM chips or any type of solid-state nonvolatile semiconductor memory) on which information is stored.
  • non-writable storage media e.g., read-only memory devices within a computer such as CD-ROM disks readable by a CD-ROM drive, ROM chips or any type of solid-state nonvolatile semiconductor memory
  • writable storage media e.g., flash memory, floppy disks within a diskette drive or hard-disk drive or any type of solid-state random-access semiconductor memory

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Computational Linguistics (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • General Engineering & Computer Science (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Artificial Intelligence (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • General Business, Economics & Management (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

To identify a product or service for which online information is to be found, input data is analysed and enhanced. The input data can be a web page retrieved by the user using a web browser on an end-user device, which is analysed by embedded code in the web page or a browser plug-in in connection with a server. Alternatively, the input data can be a reference of a particular product or service provided by the end-user device to the server in the form of a text, URL, bar code, digital image or digitized voice.

Description

Processing input data comprising references to products or services
FIELD OF THE INVENTION
The present invention relates to online searching for information about products or services. More specifically, the invention relates to processing input data comprising references to products or services.
BACKGROUND
The competitive nature of economies and the growth of the Internet have vastly increased the range of products and services available to consumers and the number of vendors of these products and services. On the Internet the proliferation of vendors and services is even faster with the presence of intermediaries such as meta-search sites and affiliate sites.
To make sure they are getting the best deals, consumer tend to use the Internet to find information and compare prices across a plurality of vendors or service providers.
For example, in the travel industry there are thousand of travel related websites offering choices of hotels, airlines and holidays. Consumers typically will only travel if they find a good deal, for which numerous web sites will be visited before making a booking.
Online publishers' content inspires users to buy products and services. To aid the navigation through web sites, publishers typically add hyperlinks to the content for products or services referenced. This is time consuming and not feasible to all forms of online content. E.g. in online forums it is substantially impossible to modify the content generated by forum users by adding hyperlinks for product or services
referenced .
From the user's perspective, linking from a referenced product or service to a single web site, which is typically an ecommerce web site, is not always the best experience. In many categories, such as travel, users expect to learn more about the product and compare prices before buying. To learn more about, for example, a hotel that appeals to them, a user typically needs to open new windows to start a search process. This prolongs a user's booking process and causes the publisher to lose their attention (as well as any potential revenue upside from a booking) .
The enormous amount of information available on the Internet makes it difficult to the consumer to navigate through the information and find the correct information. Moreover, it is difficult to compare information from various information sources without having to rely on aggregator or meta-search sites that act as intermediaries and typically offer information about preselected products and services only.
SUMMARY OF THE INVENTION
It is an object of the invention to provide for a method for processing input data comprising references to products and/or services to help consumers to find relevant information about the products and/or services.
According to an aspect of the invention a computer- implemented method in a server is proposed for processing input data. The input data comprises a reference to a product or a service. The server comprising a database. The database
comprises item identifiers and/or variants of the item
identifiers and one or more contextual references reiationally linked to the item identifiers. The item identifiers identify products and/or services. The method comprises receiving the input data from an end-user device. The method further comprises searching through the input data for a first part of the input data that matches with an item identifier or a variant of the item identifier to obtain a first matching result. The method further comprises, if the first part of the input data matched with two or more item identifiers or a variant of two or more item identifiers, searching through the input data for a second part of the input data that matches with a contextual reference to obtain a second matching result and selecting one item identifier as the first matching result based on the second matching result. The method further comprises processing based on the first matching result the first part of the input data as the reference to the product or the service. Thus, the input data provided by an end-user through his end-user device can be analysed by the server for known products and/or services. If a reference to a product or a service is found in the input data, i.e. when the first part is found, then the reference in the input data is processed to help the end-user to find relevant information about the product or service .
The embodiment of claim 2 advantageously enables a better matching of the first part with an item identifier.
The embodiment of claim 3 advantageously enables a URL, i.e. an uniform resource locator such as a web site address, to be used as input data.
The embodiment of claim 4 advantageously enables spoken input to be used as input data.
The embodiment of claim 5 advantageously enables a text to be used as input data.
The embodiment of claim 6 advantageously enables a digital image or bar code to be used as input data.
The embodiment of claim 7 advantageously enables web pages to be processed and modified to help the end-user to find relevant information about the product or service using the web page .
The embodiment of claim 8 advantageously enables limited number of references to be flagged on a web page to avoid overcrowding the webpage with visual cues.
The embodiment of claim 9 advantageously enables the generation of an online overview of information about product and services.
The embodiment of claim 10 advantageously enables the end-user to receive information directly.
The embodiment of claim 11 advantageously enables various products or services to be compared.
The embodiment of claim 12 advantageously the acquired product and service information to be consulted by the end-user at any time and on any device.
According to an aspect of the invention a computer program element is proposed, which, when being executed by a.
processor, is adapted to carry out one or more of the steps of the method above . According to an aspect of the invention a server is proposed to carry our one or more of the steps of the method above .
Hereinafter, embodiments of the invention will be described in further detail. It should be appreciated, however, that these embodiments may not be construed as limiting the scope of protection for the present invention.
BRIEF DESCRIPTION OF THE DRAWINGS
Aspects of the invention will be explained in greater detail by reference to exemplary embodiments shown in the drawings, in which:
Fig.l illustrates a flow diagram of an exemplary embodiment of the invention;
Fig.2 illustrates a network architecture of an exemplary embodiment of the invention; and
Fig 3 illustrates a flow diagram of a process for the consumer to collect and collate products and/or services of an exemplary embodiment of the invention.
DETAILED DESCRIPTION OF THE DRAWINGS
From the user's perspective, the invention enables information about products and/or services to be found more easily and enhanced. In one embodiment it becomes possible to collate the information on various related products and/or services to enable the user to compare these various products or services, e.g. to make a purchase decision. The invention enables publishers to more easily monetize their content and retain their users whilst providing users with a more convenient and more intuitive buying experience.
To identify the product or service for which online information is to be found and enhanced, input data is analysed. With reference to Fig.2, the input data can be a web page retrieved by the user using a web browser on an end-user device 20, which is analysed by embedded code in the web page or a browser plug-in in connection with a server 10. Alternatively, the input data can be a reference of a particular product or service provided by the end-user device 20 to the server 10 in the form of a text, URL, bar code, digital image or digitized voice. The user can provide the reference though a variety of communication channels 20 which may include, but are not limited to, email, text messaging, message stream, instant messaging, or an electronic interchange via an application programming interface (API) and using a variety of end-user devices which may include, but are not limited to, PC's 21, smartphones 22, tablet PC's 23, notebooks 24, set-top boxed 25 and online televisions 26.
If the input data is an URL, then the link to the URL is followed to get the referenced web page as input data. URLs in web pages may be followed as well, in which case the
referenced web page may be added to the input data.
The input data is searched for references matching an item identifier. The item identifier is a name, product id, SKU number, model number or any other unique identifier of the product or service.
The input data is parsed and matched against a
synchronised database 11 of the server 10 containing product and service item identifiers. If the input data is a text or web page, then its content is e.g. parsed and matched against a synchronised database of product and service item identifiers. If the input data is e.g. a bar code, then the bar code is matched against a synchronised database of product or service bar codes and item identifiers. If the input data is e.g. an image or an image on a web page, then the image is matched against a synchronised database of product or service images and item identifiers. If the message format is e.g. digitized voice, then the speech is converted to text and matched a synchronised database of product or service item identifiers.
The database typically contains variations of item identifiers to be found in the input data. In case of text matching, e.g. "The Sanderson Hotel", "Sanderson Hotel", "Hotel Sanderson", "The Sanderson" and "Sanderson" may all be matched to "The Sanderson Hotel". Similarly, in case of image matching, different images of the same product or service or any known advanced image recognition technology may be used to match the image to an item identifier.
The database may contain generic words, such as "The", "Hotel", "Spa", "Resort" or "Inn" and specific words, such as "Sanderson". The generic words can be detected in the input data and combined with a specific word to find the correct item identifier. E.g. a matching of the generic word "The" and/or "Hotel" near the specific word "Sanderson" results in the item identifier "The Sanderson Hotel".
References to destinations in the input data may be used to increase the confidence that the correct item identifier is found. Hereto the references to destination are stored in the database in relation to the item identifiers. The references contain e.g. countries, cities and/or street names. If e.g. in the input data the text "Ritz" is found and a reference in the form of the text "London", then the item identifier is set to the Ritz hotel in London as opposed to e.g. the Ritz hotel in New York.
Thus, the reference recognition algorithm identifies ambiguous and unambiguous phrases that are direct or indirect references to products or services on any input data using semantic rules that consider various permutations of words and symbols, and searches for a context within which they are mentioned so as to improve recognition. For example the phrase "When I visit London I like staying in the Mandarin" leads to the identification of the "Mandarin Oriental Hyde Park", the phrase "It is rumoured that Nikon is going to announce a new camera to replace the DX40" leads to the identification of the "Nikon DX-40" as opposed to e.g. the "Heathkit DX40" or the "Magnum DX40".
When an item identifier is found, the input data can be enhanced by adding information to the reference.
Fig.l shows a flow diagram of an exemplary embodiment of the invention. In step 1 the input data is received from an end-user device in the server. In step 2 the server searches through the input data for a first part of the input data that matches with an item identifier or a variant of the item
identifier to obtain a first matching result. In step 3 it is determined if the first part of the input data matched with two or more item identifiers or a variant of two or more item identifiers. If this is the case, then in step 4 the server searches through the input data for a second part of the input data that matches with a contextual reference to obtain a second matching result and selecting one item identifier as the first matching result based on the second matching result. In step 5, based on the first matching result, the first part of the input data is processed as the reference to the product or the service .
In a first scenario of the invention, the input data i an online document such as a web page wherein one or more references to products or services are to be flagged. Flagging may be done by emphasizing the reference by changing its colour and/or font face or by converting into a hyperlink to other web pages, possibly on other web sites. The online document can be user-generated or machine-generated. The references can be text- based or image-based.
The reference is recognised by comparing the text and images in the online document against a synchronised database of past and present products and services, typically within a given market segment.
The identification of the online document to be processed can be chosen either by the publisher of the document by adding program instructions in the form of embedded code to the document or by the user though the use of a browser plug-in. Examples of embedded code are JavaScript code, Flash code or any other suitable code. An example of a browser plug-in is ibid.
When a reference is matched with an item identifier In the database, the reference is flagged e.g. by adding a
hyperlink to the reference on the web page. The hyperlink is e.g. an affiliate link. The address of the link may vary
depending on the market segment or a third party promoting the trigger.
The web page, or part of the web page, is typically scraped prior to analysing by removing portions irrelevant to the content itself, such as header parts and tags.
It is typically undesirable to flag all occurrences of a reference on a web page, as this could overcrowd the web page with flags. Therefore it is determined which references should be flagged to the user and how to direct their actions based on dynamic business rules which include one or more of, but are not limited to: how many references are found; how many different references exist on the page; how many instances of a given reference appear on the page; which part(s) of the page the reference appears in; and the context of the request (e.g. time of day and location of user) .
For example, for a given publisher, if a product is referred to four times on a page, a flag may be added allowing the user to access more information about the product only for two references. For example, for a given publisher, only flag references in the main body of the page may be added, i.e. not on comments or items appearing on the side columns.
The server communicates to the user's browser the location of the targeted references on the web page the user is visiting and what action to take if the user interacts with it such as displaying on the web page additional information about the reference, providing transaction or ecommerce options, or sending the user, via a hyperlink, to another web page with related information.
Visual cues ("flags") can be added to the targeted references to call the attention of the user. Examples of visual cues are hyperlinks, colour highlight, changing text display colour or style and adding an. icon.
Upon a interaction by the user with a flagged reference, e.g. by clicking or roll-over on the reference, a preset action is performed, such as following the link or displaying additional information in a pop-up.
In a second scenario of the invention shown in Fig.3, the item identifier found in the input data is used to enable users to collect references of products or services of interest. In the second scenario consumers collect references to products or services of interest. Once collected, information about these products or services is enhanced, standardised and normalised. Tools are offered for the consumer to collate the products and services they collected and present the information.
In a first step 101 shortlist identifiers are captured. Hereto, the consumer indicates one or more products or services of interest for collecting and collating. The consumer provides the shortlist items in input data using any known message format and communication channel. The message formats may include, but are not limited to text, URLs, bar codes, images or voice. The communication channels may include, but are not limited to email, text message, instant message, electronic interchange via an application programming interface (API) .
The consumer can indicate the shortlist items directly or indirectly. Examples of a direct shortlist item are an email with the content "There are two hotels in London I am trying to decide between: the Sanderson and st martins [sic] lane" and a tweet with the content "@michel do you recommend the Sanderson". Examples of indirect shortlist items are an URL or a public social media message stream, which can be parsed for references to items consumers may want to shortlist.
In a second step 102 the item identifiers are identified in the input data. The input data is analysed to determine the relevant item identifier. If the message format is e.g. an URL, then the link is followed to get the item
identifier. If the message format is e.g. a non-linked text, then the input data is parsed and matches against a synchronised database of product or service item identifiers. If the message format is e.g. an image, then the input data is matched against a synchronised database of product or service images and item identifiers. If the message format is e.g. voice, then the input data is converted to text and matched against a synchronised, database of product or service item identifiers.
Each shortlist item may contain more than one item identifier. In this case, each item identifier has an associated accuracy probability score of how well it matches the shortlist item.
Depending on the value and distribution of the accuracy probability score, it is determined whether to select the item identifier with the highest accuracy probability score or whether to request the consumer to clarify their shortlist item.
In a third step 103 the shortlist item information is enhanced. Given the item identifier, a multitude of information may be aggregated about the shortlist items.
Sources of information may include, but are not limited to, information provided by the provider of the product or service, a synchronised database of past and present products and services, information provided by other vendors of the product or service and/or consumer reviews. The information may include hyperlinks possibly including affiliate links to other sites where the consumer can learn more about and purchase or consumption of the shortlist item, prices, features, functionality, images, user reviews, and/or any other information.
It is possible that a response from the server is provided to the user at this stage. E.g. in the example of the input data in the form of a tweet, the server may response with the tweet: "@rodrigo you can find more about the Sanderson here www. gekko . com/j 8HgT" .
In a fourth step 104 the item information is standardised, normalised and organised. Information on the products or services is standardised, normalised and organised to provide like~for~like comparison possibilities between different products or services.
In a fifth step 105 the shortlist items and all associated information are stored. The associated information includes the item identifier, the item information and any additional content contributed by the consumer and possibly by other consumers .
The additional content contributed by the consumer may include comments and annotations, indication of preferences through some form of polling and tracked usage behaviour such as views, duration of view and what content the consumers clicked on, and other information.
In a sixth step 106 the shortlist items and all associated information are presented. The consumer can view and optionally modify the shortlist items and contribute additional content through a multitude of portals or devices that include, but are not limited to, computers, computing devices mobile devices and televisions. The organisation and presentation of the item information is typically adjusted to the consumer's device capabilities.
The consumer can optionally invite other individuals, each with distinct access rights, to view and/or modify the shortlist items and contribute additional content to create a shared collaborative document.
ft server saves the chosen products and services in a list that belongs to the consumer and has customisable access rights which include but are not limited to: the consumer can choose who can see the list; the list can be private (viewable only by chosen consumers) or public (viewed by many) ; the consumer can choose who, other than themselves, can add items to the list; the consumer can choose who, other than themselves, can delete items from the list; the consumer can change any of these access rights at any time without loss of any information; others .
The invention can be used for information about any product or service. The invention can be used for the hotels market, the restaurant market, or other specific application uses .
One embodiment of the invention may be implemented as a program product for use with a computer system. The program (s) of the program product define functions of the embodiments
(including the methods described herein) and can be contained on a variety of non-transitory computer-readable storage media.
Illustrative computer-readable storage media include, but are not limited to: (i) non-writable storage media (e.g., read-only memory devices within a computer such as CD-ROM disks readable by a CD-ROM drive, ROM chips or any type of solid-state nonvolatile semiconductor memory) on which information is
permanently stored; and (ii) writable storage media (e.g., flash memory, floppy disks within a diskette drive or hard-disk drive or any type of solid-state random-access semiconductor memory) on which alterable information is stored.

Claims

1. A computer-implemented method in a server for processing input data comprising a reference to a product or a service, the server comprising a database comprising item identifiers and/or variants of the item identifiers and one or more contextual references relationally linked to the item identifiers, wherein the item identifiers identify products and/or services, the method comprising:
receiving the input data from an end-user device;
searching through the input data for a first part of the input data that matches with an item identifier or a variant of the item identifier to obtain a first matching result;
if the first part of the input data matched with two or more item identifiers or a variant of two or more item
identifiers, searching through the input data for a second part of the input data that matches with a contextual reference to obtain a second matching result and selecting one item
identifier as the first matching result based on the second matching result; and
processing based on the first matching result the first part of the input data as the reference to the product or the service .
2. The method according to claim 1, wherein the database further comprises one or more generic words
relationally linked to the item identifiers, the method further comprising:
if the first part of the input data matched with two or more item identifiers or a variant of two or more item
identifiers, searching through the input data for a third part of the input data that matches with a generic word to obtain a third matching result and selecting one item identifier as the first matching result based on the third matching result.
3. The method according to claim 1 or claim 2, wherein the input data comprises a URL and wherein the method further comprises retrieving a web page referenced by the URL and adding the web page to the input data.
4. The method according to claim 1 or claim 2, wherein the input data comprises digitized voice and wherein the method further comprises converting the digitized voice into text before searching for the first part.
5. The method according to any one of the claims 1-4, wherein the input data comprises a text and wherein the first part of the input data comprises one or more adjacent words from the text.
6. The method according to any one of the claims 1-3, wherein the input data comprises a digital image or a bar code and wherein the first part of the input data comprises the digital image or the bar code.
7. The method according to any one of the preceding claims, wherein the input data comprises an online document, wherein processing the first part comprises flagging the first part of the online document by changing the colour and/or font face of the first part, by adding an icon to the first part, or by converting the first part into a hyperlink to another web page.
3. The method according to claim 7, wherein the input data comprises two or more first parts that match with one item identifier or a variant of the one item identifier, the method further comprising:
determining which of the two or more first parts are to be processed based on one or more dynamic business rules and processing only the determined first parts.
9. The method according to any one of the claims 1-6, further comprising retrieving of information related to the first part based on the first matching result and storing the information with the first part.
10. The method according to claim 9, further comprising sending a message to the end-user device, the message comprising the information and the first part or the item identifier matching the first part.
11. The method according to claim 9, further comprising standardising, normalising and organising the first part of the input data.
12. The method according to claim 9 or claim 11, further comprising;
storing as product and service data the information and the first part or the item identifier matching the first part; and
providing the stored product and service data to the end-user device,
13. A computer program element, which, when being executed by a processor, is adapted to carry out the method according to any one of the claims 1-12.
14. A server configured to carry out the method according to any one of the claims: 1-12,
PCT/EP2011/056555 2010-04-26 2011-04-26 Processing input data comprising references to products or services WO2011134950A1 (en)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
GB1006902.9 2010-04-26
GBGB1006902.9A GB201006902D0 (en) 2010-04-26 2010-04-26 Recommendation engine
GB1008287.3 2010-05-18
GBGB1008287.3A GB201008287D0 (en) 2010-04-26 2010-05-18 Collecting and collating products on services

Publications (1)

Publication Number Publication Date
WO2011134950A1 true WO2011134950A1 (en) 2011-11-03

Family

ID=42270801

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/EP2011/056555 WO2011134950A1 (en) 2010-04-26 2011-04-26 Processing input data comprising references to products or services

Country Status (2)

Country Link
GB (2) GB201006902D0 (en)
WO (1) WO2011134950A1 (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2005098680A2 (en) * 2004-03-31 2005-10-20 Google Inc. Query rewriting with entity detection
WO2006065583A1 (en) * 2004-12-14 2006-06-22 Google Inc. Providing useful information associated with an item in a document
WO2006065546A2 (en) * 2004-12-14 2006-06-22 Google, Inc. Method, system and graphical user interface for providing reviews for a product

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2005098680A2 (en) * 2004-03-31 2005-10-20 Google Inc. Query rewriting with entity detection
WO2006065583A1 (en) * 2004-12-14 2006-06-22 Google Inc. Providing useful information associated with an item in a document
WO2006065546A2 (en) * 2004-12-14 2006-06-22 Google, Inc. Method, system and graphical user interface for providing reviews for a product

Also Published As

Publication number Publication date
GB201008287D0 (en) 2010-06-30
GB201006902D0 (en) 2010-06-09

Similar Documents

Publication Publication Date Title
US9390144B2 (en) Objective and subjective ranking of comments
US9070088B1 (en) Determining trustworthiness and compatibility of a person
US8352455B2 (en) Processing a content item with regard to an event and a location
US10685181B2 (en) Linguistic expression of preferences in social media for prediction and recommendation
US8812505B2 (en) Method for recommending best information in real time by appropriately obtaining gist of web page and user's preference
US8150979B1 (en) Supporting multiple landing pages
KR20140016247A (en) Enabling advertisers to bid on abstract objects
US9104746B1 (en) Identifying contrarian terms based on website content
CN112486917A (en) Method and system for automatically generating information-rich content from multiple microblogs
US20130117716A1 (en) Function Extension for Browsers or Documents
CN111052109B (en) Expert search thread invitation engine
US9886711B2 (en) Product recommendations over multiple stores
KR101981136B1 (en) Program, method and system for providing recommendation contents
US20160299951A1 (en) Processing a search query and retrieving targeted records from a networked database system
KR20130011557A (en) System and method for providing automatically completed query by regional groups
Menk et al. Recommendation systems for tourism based on social networks: A survey
KR100964090B1 (en) Method and system for recommending advertisement keyword by analyzing log
US20140214541A1 (en) Method and system for user-controlled rendering of mobile advertisements
KR101132431B1 (en) System and method for providing interest information
KR20130100841A (en) Apparatus and method for decideing target advertisingusing interest information of user
AlMeshary et al. A recommendation system for Twitter users in the same neighborhood
WO2011134950A1 (en) Processing input data comprising references to products or services
KR101347220B1 (en) System for providing advertisement contents
KR101407207B1 (en) Database server for categorizing/offering recommendation item by the category and method thereof
KR101591716B1 (en) Advertisement providing system and mehtod for restricting buyable keyword to efficiently provide on-line local avertisement using address and type of business

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 11718691

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 11718691

Country of ref document: EP

Kind code of ref document: A1