US20170316088A9 - Methods and systems for modifying a user profile for a recommendation algorithm and making recommendations based on user interactions with items - Google Patents

Methods and systems for modifying a user profile for a recommendation algorithm and making recommendations based on user interactions with items Download PDF

Info

Publication number
US20170316088A9
US20170316088A9 US14046229 US201314046229A US2017316088A9 US 20170316088 A9 US20170316088 A9 US 20170316088A9 US 14046229 US14046229 US 14046229 US 201314046229 A US201314046229 A US 201314046229A US 2017316088 A9 US2017316088 A9 US 2017316088A9
Authority
US
Grant status
Application
Patent type
Prior art keywords
user
item
user profile
modified
profile
Prior art date
Legal status (The legal status 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 status listed.)
Pending
Application number
US14046229
Other versions
US20140108395A1 (en )
Inventor
Nir Polonsky
Girish Malangi
Kevin Volpe
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Gartner Inc
Original Assignee
Gartner Inc
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

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/30Information retrieval; Database structures therefor ; File system structures therefor
    • G06F17/3061Information retrieval; Database structures therefor ; File system structures therefor of unstructured textual data
    • G06F17/30699Filtering based on additional data, e.g. user or group profiles
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/30Information retrieval; Database structures therefor ; File system structures therefor
    • G06F17/30017Multimedia data retrieval; Retrieval of more than one type of audiovisual media
    • G06F17/30023Querying
    • G06F17/30029Querying by filtering; by personalisation, e.g. querying making use of user profiles
    • G06F17/30035Administration of user profiles, e.g. generation, initialisation, adaptation, distribution
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/30Information retrieval; Database structures therefor ; File system structures therefor
    • G06F17/3061Information retrieval; Database structures therefor ; File system structures therefor of unstructured textual data
    • G06F17/30699Filtering based on additional data, e.g. user or group profiles
    • G06F17/30702Profile generation, learning or modification
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/30Information retrieval; Database structures therefor ; File system structures therefor
    • G06F17/30943Information retrieval; Database structures therefor ; File system structures therefor details of database functions independent of the retrieved data type
    • G06F17/30964Querying
    • G06F17/30967Query formulation
    • G06F17/3097Query formulation using system suggestions

Abstract

Methods and apparatus for modifying a user profile for a recommendation algorithm are provided. A user is provided with electronic access to an item. The item may comprise one of a document, an article, a chart, a graphic, a report, a web page, or the like. User interaction with the item is enabled. The user interaction with the item is then electronically tracked and stored. The user's user profile used by a recommendation engine is then modified based on the tracked user interactions. The user interaction may comprise at least one of annotating, highlighting, modifying, customizing, adding comments to the item, and the like. The user modified item can be saved and details of the user interaction with the item may be used to modify the user profile. At least one of items or peer recommendations can then be provided to the user based on the modified user profile.

Description

  • This application claims the benefit of U.S. provisional patent application no. 61/712,470 filed on Oct. 11, 2012, which is incorporated herein and made a part hereof by reference.
  • BACKGROUND OF THE INVENTION
  • The present invention relates to the field of item and peer recommendation algorithms directed towards industry professionals. More specifically, the present invention is aimed at providing a more accurate and iterative recommendation algorithm that is adaptable based on a continuum of user actions or inputs rather than based on binary yes/no signals.
  • Prior art recommendation algorithms typically rely on a user profile and a set of filters for matching the user profile to an item or object (such as a document, a web page, a presentation, an article, a chart or graphic, a product, or the like) or a peer (such as an expert, a colleague in the same or similar industry, a service provider, or the like).
  • In item-to-item, item-to-user and user-to-user recommendation algorithms, one of the biggest challenges is to profile users accurately so that the right set of recommendations can be made. Traditionally recommendation engines have been able to use a user's declared or explicit profile and also the user's web site behavioral profile. The user's explicit profile includes the information he provides; for example, during a registration process, whereas the user's implicit profile may include the user's web site behavior such as the key words he or she searched on, articles he or she buys, articles he or she looks at, peer discussions he or she participates in, and the like. Such prior art recommendation engines are based on a user's set profile and simple binary scoring of the users actions indicating whether an item was viewed/not viewed or used/not used (e.g., in the form of a 1/0 or Y/N score), which records are then smoothed via weightings and/or time decay functions for use in the recommendation engine.
  • It would be advantageous to be able to adapt a user's profile iteratively based on the continuous spectrum of the user's interactions with different items, so that the recommendations can be more specifically tailored to a particular user. It would be particularly advantageous if such adaptation of the user's profile could be based on feedback obtained from a user, including feedback based on a user's interaction with a recommended item or direct input from a user as reflected in their modification of an item.
  • The methods, apparatus, and systems of the present invention provide the foregoing and other advantages.
  • SUMMARY OF THE INVENTION
  • The present invention relates to the field of item and peer recommendation algorithms directed towards industry professionals. More particularly, the present invention relates to methods and apparatus for modifying a user profile for a recommendation algorithm in order to more specifically tailor content based on a user's interactions with previously recommended content. Methods and systems for providing recommendations using such a modifiable user profile are also provided.
  • In accordance with one example embodiment of the present invention, a computerized method for modifying a user profile for a recommendation algorithm is provided. In such an example embodiment, a user is provided with electronic access to an item. The item may comprise one of a document, an article, a chart, a graphic, a report, a web page, or the like. User interaction with the item is enabled. The user interaction with the item is then electronically tracked and stored. The user's user profile used by a recommendation engine is then modified based on the tracked user interactions.
  • The user interaction may comprise at least one of annotating, highlighting, modifying, customizing, adding comments to the item, and the like.
  • The user interaction may be enabled via an application service provider application or a downloadable software application.
  • A web page displaying the item may be enabled to be displayed as an interactive web page providing capabilities for at least one of entering annotations, entering comments, adding highlighting of text portions, making modifications to the item, and the like.
  • In one example embodiment, the method may further comprise saving the user modified item, recording details of the user interaction from the user modified item, and using the details of the user interaction to modify the user profile.
  • The details of the user interaction may comprise at least one of: a coordinate position of the annotation, modification, comment, or highlighting in the item; the content of the annotation, modification, comment, and highlighting; meta data describing the content being annotated, modified, commented on, or highlighted; meta data assigned to the annotation, modification, comment, or highlighting; results of sentiment analysis of the annotation, modification, comment, or highlighted content, and the like.
  • The user profile may be modified based on at least one of: key words taken from the annotation, modification, comment, or highlighted content; meta data describing the annotation, modification, comment; key words from the highlighted content; meta data of a highlighted section; meta data describing the item being modified in an info-graphic interface, and the like.
  • The user profile may be stored in a database and modified at least one of: periodically at defined intervals; after each user interaction; after a defined number of user interactions; after a number of user interactions supersede a predetermined threshold number of interactions, and the like.
  • The user may access the item via a web site, a mobile application, an interactive email, or the like. Alternatively, the item may be provided to the user via the recommendation engine, or the item may be accessed from a database.
  • In another example embodiment, the item may comprise an info-graphics chart with weighted data elements. The user interaction may comprise modification of one or more weights assigned to the data elements.
  • The info-graphics chart may comprise an X axis and a Y axis. The X axis and the Y axis may each correspond to weighted sums of sub scores along multiple sub-dimensions. Weights may be assigned to each sub-dimension. User modification of the weight for a particular sub-dimension may indicate a relative priority of the particular sub-dimension to the user.
  • User modification of the one or more weights may be enabled via an interactive graphics display comprising one of a slidebar for each of the weights, a clickable weight scale for selecting the desired weight for each sub-dimension, a user interface embedded in a web page providing the info-graphic, an editable table listing the weights for each sub-dimension, and the like.
  • At least one of items or peer recommendations may be provided to the user based on the modified user profile. Further, the items or peer recommendations provided to the user may be modified based on the modified user profile. For example, modifications to the items or peer recommendations may comprise at least one of highlighting of key words in a recommended document, creating a custom view in an info-graphics chart, re-ordering of text elements in the recommended document, inserting annotations into an item, cropping or customizing of the info-graphics chart, providing customized contextualized comments to the recommendations which relates the recommendations to the modified user profile, re-ordering a list of recommendations, or the like.
  • The user profile in an unmodified state may comprise at least one of an explicit profile comprising data entered by the user and an implicit profile comprising information obtained from user behavior. The user behavior may comprise at least one of key words used in key word searches, articles reviewed, web pages reviewed, articles purchased, discussions reviewed, discussions participated in, peer profiles selected for connection, articles or web pages saved or downloaded, items clicked on in an info-graphics chart, and the like.
  • In accordance with the present invention, an apparatus for modifying a user profile for a recommendation algorithm is also provided. The apparatus may comprise a user interface adapted to provide a user with electronic access to an item and adapted to enable user interaction with the item. The item may comprise one of a document, an article, a chart, a graphic, a report, a web page, or the like. The apparatus may also comprise a software application for electronically tracking the user interaction with the item, a database for storing the user interaction with the item, and a database for storing a user profile. The software application is adapted to modify the user profile based on the tracked user interactions. The modified user profile may then be used by a recommendation engine to serve items or peer recommendations to the user tailored to the modified user profile.
  • The above-described apparatus may also include or enable additional features discussed above in connection with the various embodiments of the methods for modifying a user profile.
  • In addition, the present invention also encompasses methods and systems for providing recommendations using a continuously modifiable user profile. In one example embodiment of such a method, an initial user profile is entered and a user is provided with electronic access to an item. The item may comprise one of a document, an article, a chart, a graphic, a report, a web page, or the like. User interaction with the item is enabled. The user interaction with the item is electronically tracked and stored. The initial user profile may then be modified based on the tracked user interactions. One or more items or peer recommendations can then be served to the user based on the modified user profile.
  • In an example embodiment of a system for providing recommendations using a continuously modifiable user profile, the system may comprise a user interface providing a user with electronic access to an item and enabling user interaction with the item. The item may comprise one of a document, an article, a chart, a graphic, a report, a web page, or the like. A software application is provided for electronically tracking the user interaction with the item. A database is provided for storing the user interaction with the item. A database is also provided for storing a user profile. A recommendation engine is provided for serving items to users based on the user profile. The software application modifies the user profile based on the tracked user interactions. The modified user profile is used by a recommendation engine to serve one or more items or peer recommendations to the user tailored to the modified user profile.
  • The one or more items or peer recommendations served to the user may be at least one of modified and customized by the recommendation engine based on the modifications to the user profile.
  • Such methods and systems for providing recommendations may implement the various embodiments of the methods for modifying a user profile discussed above.
  • BRIEF DESCRIPTION OF THE DRAWINGS
  • The present invention will hereinafter be described in conjunction with the appended drawing figures, wherein like reference numerals denote like elements, and:
  • FIG. 1 shows a block diagram of an example embodiment of the present invention;
  • FIG. 2A shows an example of an item loaded on an interactive web page in original form;
  • FIG. 2B shows the item of FIG. 2A with a pop-up menu which facilitates annotations and highlighting of the item;
  • FIG. 2C shows the item of FIG. 2A with highlighting added;
  • FIG. 2D shows the item of FIG. 2A with a note being added;
  • FIG. 2E shows the item of FIG. 2A with an aggregation function;
  • FIG. 3 shows an example of an info-graphics chart with user modifiable weighting functions in accordance with an example embodiment of the present invention; and
  • FIG. 4 shows a flow diagram of an example embodiment of the present invention.
  • DETAILED DESCRIPTION
  • The ensuing detailed description provides exemplary embodiments only, and is not intended to limit the scope, applicability, or configuration of the invention. Rather, the ensuing detailed description of the exemplary embodiments will provide those skilled in the art with an enabling description for implementing an embodiment of the invention. It should be understood that various changes may be made in the function and arrangement of elements without departing from the spirit and scope of the invention as set forth in the appended claims.
  • The present invention relates to the provision of the ability to continuously update, modify, supplement, and/or customize a user profile for use in a recommendation algorithm based on a user's interaction with an item or object that is either provided by the recommendation algorithm to the user or available to them otherwise, provided that the interaction of such item can be tracked or monitored (whether in real-time or subsequently) by the recommendation system. The user can also be provided with the ability to directly modify and customize certain elements of his/her profile, for example, by changing or modifying various weighting factors or reordering elements, as will be explained in more detail below.
  • It should be appreciated that the term “user” as used herein may be interpreted to mean an individual, a company, an association, an organization, a collection of individuals, or the like.
  • One example of a recommendation engine is disclosed in commonly-owned co-pending U.S. patent application Ser. No. 13/528,029 filed on Jun. 20, 2012 entitled Bi-Model Recommendation Engine for Recommending Items and Peers, which is respectfully incorporated herein and made a part hereof by reference. The present invention can be used in connection with the recommendation engine described in U.S. patent application no. Ser. No. 13/528,029.
  • In one example embodiment of the present invention, a computerized method for modifying a user profile for a recommendation algorithm is provided. FIG. 1 shows an example embodiment of a system for implementing such a method. In one example embodiment, a user is provided with electronic access to an item 17 (e.g., via an Internet-enabled device 10 with a user interface 12). The item 17 may comprise one of a document, an article, a chart, a graphic, a report, a web page, a business document, or any type of modifiable electronic information item.
  • The user may access the item 17 via a web site, a mobile application, an interactive email, or the like via the Internet-enabled device 10. Alternatively, the item 17 may be provided to the user via a recommendation engine 14, or the item 17 may be accessed from a database 16 or other trackable source (e.g., remote server 30 or other Internet source).
  • Those skilled in the art should appreciate that the connections between the elements of FIG. 1 may be enabled via the Internet, and that the Internet enabled device 10 may be used to obtain items from other sources via the Internet, which items can then be modified by the user and processed by the system as discussed herein.
  • User interaction with the item 17 is enabled. The user interaction with the item 17 is then electronically tracked and stored (e.g., via an interaction database 18 and associated tracking software 19). The user's user profile 21 (e.g., stored in a profile database 20) used by the recommendation engine 14 may then be modified based on the tracked user interactions. Although the tracking software 19 is shown in FIG. 1 as part of the interaction database 18, those skilled in the art will appreciate that it can be implemented separately therefrom as long as it is in communication with both the interaction database 18 and profile database 20.
  • The user interaction may comprise at least one of annotating, highlighting, modifying, customizing, adding comments electronically within the item 17, or otherwise personalizing the item 17. The user interaction may be enabled via an application service provider application or a downloadable software application resident on the Internet-enabled device 10. The Internet enabled device 10 may comprise a computer, laptop, personal computing device, tablet, e-book reader, smart phone, or the like.
  • A web page displaying the item 17 may be enabled to be displayed as an interactive web page providing capabilities for at least one of entering annotations, entering comments, adding highlighting of text portions, making modifications to the item, and the like. For example, when viewing the item, the user may click on an “annotate” icon, which causes the page being viewed to be displayed as an interactive page with editorial functions consistent with the present invention, including highlighting, annotating, modifying, customizing, commenting, and the like. FIGS. 2A-2E illustrate one example embodiment of modifying an item displayed as an interactive web page. FIG. 2A shows an original item, which in the example shown is an article 22, in an unmodified state. FIG. 2B shows the article 22 with a pop-up menu 24 once text is selected (e.g., via highlighting with a mouse or similar tool), which enables either insertion of permanent highlighting or annotating. FIG. 2C shows the article 22 with permanent highlighting 26 added by the user. FIG. 2D shows the article 22 with a note 28 being added by the user. FIG. 2E shows the article 22 with a aggregation function 29 that aggregates all modifications made to the article 22 (which in the example shown includes the highlighted text and corresponding note) and allows for additional notes regarding the article 22 in general.
  • In one example embodiment, the method may further comprise saving the user modified item 17 (e.g., in database 16), recording details of the user interaction from the user modified item 17 (e.g., in interaction database 18), and using the details of the user interaction to modify the user profile 21, which may be stored in the profile database 20.
  • The details of the user interaction stored in the interaction database 18 may comprise at least one of: a coordinate position of the annotation, modification, comment, or highlighting in the item; the content of the annotation, modification, comment, and highlighting; meta data describing the content being annotated, modified, commented on, or highlighted; meta data assigned to the annotation, modification, comment, or highlighting; results of sentiment analysis of the annotation, modification, comment, or highlighted content, and any similar methods which capture the essence of the modification.
  • The user profile 21 may be modified based on at least one of: key words taken from the annotation, modification, comment, or highlighted content; meta data describing the annotation, modification, comment; key words from the highlighted content; meta data of a highlighted section; meta data describing the item being modified in an info-graphic interface, and the like.
  • Alternatively, once the user has completed reviewing and modifying the item 17, the annotated item 17 is saved on a remote server 30 (rather than database 16) for future use (both by the user and by the recommendation engine 14). The remote server 30 may then record the annotation details as discussed above and save this information under a user ID. The user can also save the item locally (on the Internet-enabled device 10 or a storage device associated therewith) as a PDF file for printing and sharing purposes. In such an embodiment, the remote server 30 may also include tracking software 19 and be in communication with both the recommendation engine 14 and the profile database 20, such that the user profiles can be updated in the same manner as discussed above. It should be appreciated that, although the recommendation engine 14 and other components shown in FIG. 1 are shown as separate functional units, any one or more of the components shown in FIG. 1 (other than the Internet enabled device 10 and user interface 12) can be consolidated into one functional unit. For example, the recommendation engine 14, the database 16, the interaction database 18, and/or the profile database 20 can be implemented as part of the remote server 30. It is also contemplated that a document or other item may be annotated “off line” apart from the system or the corresponding software application, and later uploaded to the system and/or stored at the database 16 or server 30. The annotation details can then be processed as discussed above.
  • The user profile 21 may be stored in the profile database 20 and modified at least one of: periodically at defined intervals; after each user interaction; after a defined number of user interactions; after a number of user interactions supersede a predetermined threshold number of interactions, and the like. Software 15 for use in modifying the user profile may be provided in the profile database 20. Although the profile modification software 15 is shown in FIG. 1 as part of the profile database 20, those skilled in the art will appreciate that it can be implemented separately therefrom and be in communication with the interaction database 18 and/or the tracking software 19 and the profile database 20.
  • In another example embodiment as shown in FIG. 3, the item 17 may comprise an info-graphics chart 32 with weighted data elements 33. The user interaction may comprise modification of one or more weights 34 assigned to the data elements 33. Changing weights assigned to the data elements 33 in the chart 32 results in a change in the element position in the chart 32.
  • As shown in FIG. 3, the info-graphics chart may comprise an X axis and a Y axis. The X axis and the Y axis may each correspond to weighted sums of sub scores along multiple sub-dimensions. Weights may be assigned to each sub-dimension. User modification of the weight 34 for a particular sub-dimension may indicate a relative priority of the particular sub-dimension to the user.
  • User modification of the one or more weights may be enabled in a variety of ways. For example, as shown in FIG. 3, user modification of the one or more weights may be enabled via an interactive graphics display comprising a slidebar 36 for each of the weights, or alternatively, via one of a clickable weight scale for selecting the desired weight for each sub-dimension, a user interface embedded in a web page providing the info-graphic, an editable table listing the weights for each sub-dimension, or any similar info-graphics methods for modifying weights and/or scores. The starting position of the graphic (scale, chart, etc.) represents the system assigned weightings.
  • FIG. 3 shows the example info-graphics chart 32 as the Magic Quadrant chart developed by Gartner, Inc. The Magic Quadrant chart depicts markets in the middle phases of their life cycle by using a two-dimensional matrix that evaluates vendors based on their completeness of vision and ability to execute. The Magic Quadrant uses weighted criteria 34 that plot vendors based on their relative strengths in the market. This Gartner, Inc. model is well suited for high-growth and consolidating markets where market and vendor differentiations are distinct. For example, the coordinate of a map element (e.g., vendor position in a quadrant (represented by data elements 33)) is determined by a calculated X and Y score which are by themselves a weighted sum of sub-scores along multiple sub dimensions. The X and Y scores may relate to a vendors completeness of vision and ability to execute. The completeness of vision score may be based on the following sub-dimensions: the vendor's market understanding, marketing strategy, sales strategy, offering or product strategy, business model, vertical/industry strategy, innovation, geographic strategy, and the like. The ability to execute score may be based on the following sub-dimensions: the vendor's products or services, overall visibility, sales execution and pricing, market responsiveness, track record, marketing execution, customer experience, operations, and the like. Other sub-dimension factors may of course be considered for each. Weightings may be provided for the sub-dimensions based on the importance of each factor in the overall score as determined by Gartner (or other provider of an info-graphics chart).
  • While the above-embodiments of present invention are described herein in connection with Gartner's Magic Quadrant, those skilled in the art should appreciate that it can be applied to any info-graphics chart from any provider, as well as to any type of business metric, graph, table, or the like where weighting factors and/or scores are used.
  • For example, with the present invention, the user may receive a recommendation to access, or otherwise accesses or views, a Magic Quadrant or other info-graphics chart (e.g., via an interactive web page or a software application), which provides the Gartner view with the Gartner calculated coordinates, and the user is allowed to create a custom view by modifying the weights 34 for each of the sub dimensions, and hence, recalculating vendor positions along X and Y coordinates. For example, the user may feel that the sub-dimension “sales execution and pricing” is of critical importance, and increase the weighting factor applied to this sub-dimension, resulting in vendors having a higher score in this area being given an upgraded position in the Magic Quadrant. The system then records and saves the customized view in the user's extended profile or a similar table, which can then be applied to further recommendations for that user.
  • It should be appreciated that access to the info-graphics, such as the Magic Quadrant, can be provided in a variety of ways, including but not limited to via an interactive web page or a software application (either provided remotely via an application service provider or a downloadable application resident on a computer, laptop, personal computing device, tablet, e-book reader, smart phone, or the like).
  • Once the user has adjusted the weightings 34, the resultant custom view of the info-graphics is saved for future use, for example on a database 16 or remote server 30. The database 16 or remote server 30 records the custom weight table for the particular user, for example under an info-graphic ID assigned to that user as part of the user profile 21 (as described above).
  • It is also contemplated that the info-graphics may be modified “off line” apart from the system or the corresponding software application, and later uploaded to the system and/or stored at the database 16 or server 30. The changes in the weighting details can then be processed as discussed above.
  • At least one of items or peer recommendations may be provided from the recommendation engine 14 to the user based on the modified user profile 21. Further, the items or peer recommendations provided to the user may be modified based on the modified user profile 21. For example, modifications to the items or peer recommendations may comprise at least one of highlighting of key words in a recommended document, creating a custom view in an info-graphics chart, re-ordering of text elements in the recommended document, inserting annotations into an item, cropping or customizing of the info-graphics chart, providing customized contextualized comments to the recommendations which relates the recommendations to the modified user profile, re-ordering a list of recommendations, or the like.
  • The user profile 21 in an unmodified state may comprise at least one of an explicit profile comprising data entered by the user and an implicit profile comprising information obtained from user behavior. The user behavior may comprise at least one of key words used in key word searches, articles reviewed, web pages reviewed, articles purchased, discussions reviewed, discussions participated in, peer profiles selected for connection, articles or web pages saved or downloaded, items clicked on in an info-graphics chart, and the like.
  • In addition, the present invention also encompasses methods and systems for providing recommendations using a continuously modifiable user profile. FIG. 4 shows a flow diagram of one example embodiment of such a method. An initial user profile 21 is entered and a user is provided with electronic access to an item 17 via the recommendation engine 14. The user profile 21 may include a user ID 41 and other profile information 42. To enable modification of the user profile 21, the user profile may have the capability to maintain modifiable fields relating to profile attributes (such as meta data or other information used to modify the user profile 21 as discussed above) as well as recency of any such modifications (e.g., such as a time stamp of the most recent modification to the user profile 21). As discussed above, the item 17 may comprise one of a document, an article, a chart, a graphic, a report, a web page, or any type of modifiable electronic information item. User interaction with the item is enabled as discussed above in connection with FIG. 1. For example, the user may initiate a user session 45 with the recommendation engine 14. The user interaction with the item 17 is electronically tracked and stored in the interaction database 18 (e.g., via associated tracking software 19 shown in FIG. 1). The interaction database 18 may also track and store a time stamp for the modification to each item, the user ID, the item ID for each item, and the modification meta data (or other details of the interaction as discussed above). The initial user profile 21 may then be modified based on the tracked user interactions. For example, the profile modification software 15 (shown in FIG. 1) may modify the attributes field 43 and the recency field 44 of the user profile 21 based on the received interaction details. One or more items 17 or peer recommendations can then be served to the user based on the modified user profile 21.
  • The one or more items 17 or peer recommendations served to the user may be at least one of modified and customized by the recommendation engine 14 based on the modifications to the user profile 21. For example, it can highlight key words in a new item it serves in advance, based on the user's recorded recent highlights. In another example, the recommendation engine may also create a custom view in a new Magic Quadrant it serves based on the user's recently recorded custom weights. Additional modifications to recommended items may also be provided by the recommendation engine based on the modified user profile, such as re-ordering of text or elements of an item, inserting annotations into an item that relate to a user's comments or annotations in a recently modified item, cropping and/or customizing info-graphics based on user's customization of similar graphics, as well as other conceivable modifications.
  • Such methods and systems for providing recommendations may implement the various embodiments of the methods for modifying a user profile discussed above.
  • It should be appreciated that the process described above in connection with FIG. 4 is iterative such that the user profile is continuously updated based on user interactions with items over time (e.g., over all user sessions 45) so that the items and recommendations served by the recommendation engine are continuously tailored to the user and the users evolving preferences and interests.
  • With the present invention, the user's interests and priorities at a given point of time are implicitly and explicitly derived from the richness provided by the spectrum of the user's recent interactions with, and modifications of, an information item or items. Inputs such as annotations, highlights, saved comments, and modifications of info-graphics, provide valuable data about the user's interests and priorities beyond the value provided by the binary signals already available (for example, coordinates of highlighted paragraphs vs. a binary signal “read the document=yes”). This data, once aggregated, tabulated, prioritized (by date, significance, depth of interaction, etc.), and mapped to the relevant taxonomy, can be used in order to calculate to time-sensitive relevancy of related information items (articles, charts, peer profiles, event announcements, news items, etc.) to be recommended to a given user at a given time point. It can also be used to define user clusters to be used in collaborative filtering and other recommendation logic.
  • It should now be appreciated that the present invention provides advantageous methods and systems for providing more targeted items and peer recommendations from a recommendation algorithm. In particular, it should be appreciated that the present invention moves away from the limited binary scoring of prior art recommendation engines to a more iterative approach based on a continuum of inputs from the user's interaction with and/or modifications to an item (rather than simply whether an item was used/viewed). Although the invention has been described in connection with various illustrated embodiments, numerous modifications and adaptations may be made thereto without departing from the spirit and scope of the invention as set forth in the claims.

Claims (22)

    What is claimed is:
  1. 1. A computerized method for modifying a user profile for a recommendation algorithm, comprising:
    providing a user with electronic access to an item, the item comprising one of a document, an article, a chart, a graphic, a report, and a web page;
    enabling user interaction with the item;
    electronically tracking and storing the user interaction with the item; and
    modifying the user's user profile used by a recommendation engine based on the tracked user interactions.
  2. 2. A method in accordance with claim 1, wherein the user interaction comprises at least one of annotating, highlighting, modifying, customizing, and adding comments to the item.
  3. 3. A method in accordance with claim 2, wherein the user interaction is enabled via an application service provider application or a downloadable software application.
  4. 4. A method in accordance with claim 2, further comprising:
    enabling a web page displaying the item to be displayed as an interactive web page providing capabilities for at least one of entering annotations, entering comments, adding highlighting of text portions, and making modifications to the item.
  5. 5. A method in accordance with claim 4, further comprising:
    saving the user modified item;
    recording details of the user interaction from the user modified item;
    using the details of the user interaction to modify the user profile.
  6. 6. A method in accordance with claim 5, wherein the details of the user interaction comprise at least one of: a coordinate position of the annotation, modification, comment, or highlighting in the item; the content of the annotation, modification, comment, and highlighting; meta data describing the content being annotated, modified, commented on, or highlighted; meta data assigned to the annotation, modification, comment, or highlighting; and results of sentiment analysis of the annotation, modification, comment, or highlighted content.
  7. 7. A method in accordance with claim 6, wherein the user profile is modified based on at least one of: key words taken from the annotation, modification, comment, or highlighted content; meta data describing the annotation, modification, or comment; key words from the highlighted content; meta data of a highlighted section; and meta data describing the item being modified in an info-graphic interface.
  8. 8. A method in accordance with claim 1, wherein the user profile is stored in a database and modified at least one of: periodically at defined intervals; after each user interaction; after a defined number of user interactions; and after a number of user interactions supersede a predetermined threshold number of interactions.
  9. 9. A method in accordance with claim 1, wherein one of: the user accesses the item via a web site, a mobile application, or interactive email; the item is provided to the user via the recommendation engine; and the item is accessed from a database.
  10. 10. A method in accordance with claim 1, wherein:
    the item comprises an info-graphics chart with weighted data elements;
    the user interaction comprises modification of one or more weights assigned to the data elements.
  11. 11. A method in accordance with claim 10, wherein:
    the info-graphics chart may comprise an X axis and a Y axis;
    the X axis and the Y axis each correspond to weighted sums of sub scores along multiple sub-dimensions;
    weights are assigned to each sub-dimension;
    user modification of the weight for a particular sub-dimension indicates a relative priority of the particular sub-dimension to the user.
  12. 12. A method in accordance with claim 11, wherein:
    user modification of the one or more weights is enabled via an interactive graphics display comprising one of a slidebar for each of the weights, a clickable weight scale for selecting the desired weight for each sub-dimension, a user interface embedded in a web page providing the info-graphic, and an editable table listing the weights for each sub-dimension.
  13. 13. A method in accordance with claim 1, further comprising:
    providing at least one of items or peer recommendations to the user based on the modified user profile.
  14. 14. A method in accordance with claim 13, wherein items or peer recommendations provided to the user are modified based on the modified user profile.
  15. 15. A method in accordance with claim 14, wherein modifications to the items or peer recommendations comprise at least one of highlighting of key words in a recommended document, creating a custom view in an info-graphics chart, re-ordering of text elements in the recommended document, inserting annotations into an item, cropping or customizing of the info-graphics chart, providing customized contextualized comments to the recommendations which relates the recommendations to the modified user profile, and re-ordering a list of recommendations.
  16. 16. A method in accordance with claim 1, wherein the user profile in an unmodified state comprises at least one of an explicit profile comprising data entered by the user and an implicit profile comprising information obtained from user behavior.
  17. 17. A method in accordance with claim 16, wherein the user behavior comprises at least one of key words used in key word searches, articles reviewed, web pages reviewed, articles purchased, discussions reviewed, discussions participated in, peer profiles selected for connection, articles or web pages saved or downloaded, and items clicked on in an info-graphics chart.
  18. 18. An apparatus for modifying a user profile for a recommendation algorithm, comprising:
    a user interface providing a user with electronic access to an item, the item comprising one of a document, an article, a chart, a graphic, a report, and a web page, and enabling user interaction with the item;
    a software application for electronically tracking the user interaction with the item;
    a database for storing the user interaction with the item;
    a database for storing a user profile;
    wherein:
    the software application modifies the user profile based on the tracked user interactions; and
    the modified user profile is used by a recommendation engine to serve items or peer recommendations to the user tailored to the modified user profile.
  19. 19. A method for providing recommendations using a continuously modifiable user profile, comprising:
    entering an initial user profile;
    providing a user with electronic access to an item, the item comprising one of a document, an article, a chart, a graphic, a report, and a web page;
    enabling user interaction with the item;
    electronically tracking and storing the user interaction with the item;
    modifying the initial user profile based on the tracked user interactions; and
    serving one or more items or peer recommendations to the user based on the modified user profile.
  20. 20. A method in accordance with claim 19, wherein the one or more items or peer recommendations served to the user are at least one of modified and customized by the recommendation engine based on the modifications to the user profile.
  21. 21. A system for providing recommendations using a continuously modifiable user profile, comprising:
    a user interface providing a user with electronic access to an item, the item comprising one of a document, an article, a chart, a graphic, a report, and a web page and enabling user interaction with the item;
    a software application for electronically tracking the user interaction with the item;
    a database for storing the user interaction with the item;
    a database for storing a user profile;
    a recommendation engine for serving items to users based on the user profile;
    wherein:
    the software application modifies the user profile based on the tracked user interactions; and
    the modified user profile is used by a recommendation engine to serve one or more items or peer recommendations to the user tailored to the modified user profile.
  22. 22. A system in accordance with claim 20, wherein the one or more items or peer recommendations served to the user are at least one of modified and customized by the recommendation engine based on the modifications to the user profile.
US14046229 2010-02-03 2013-10-04 Methods and systems for modifying a user profile for a recommendation algorithm and making recommendations based on user interactions with items Pending US20170316088A9 (en)

Priority Applications (4)

Application Number Priority Date Filing Date Title
US12658192 US20110191311A1 (en) 2010-02-03 2010-02-03 Bi-model recommendation engine for recommending items and peers
US13528029 US8661034B2 (en) 2010-02-03 2012-06-20 Bimodal recommendation engine for recommending items and peers
US201261712470 true 2012-10-11 2012-10-11
US14046229 US20170316088A9 (en) 2010-02-03 2013-10-04 Methods and systems for modifying a user profile for a recommendation algorithm and making recommendations based on user interactions with items

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
US14046229 US20170316088A9 (en) 2010-02-03 2013-10-04 Methods and systems for modifying a user profile for a recommendation algorithm and making recommendations based on user interactions with items

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
US13528029 Continuation-In-Part US8661034B2 (en) 2010-02-03 2012-06-20 Bimodal recommendation engine for recommending items and peers

Publications (2)

Publication Number Publication Date
US20140108395A1 true US20140108395A1 (en) 2014-04-17
US20170316088A9 true true US20170316088A9 (en) 2017-11-02

Family

ID=50476369

Family Applications (1)

Application Number Title Priority Date Filing Date
US14046229 Pending US20170316088A9 (en) 2010-02-03 2013-10-04 Methods and systems for modifying a user profile for a recommendation algorithm and making recommendations based on user interactions with items

Country Status (1)

Country Link
US (1) US20170316088A9 (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9477672B2 (en) 2009-12-02 2016-10-25 Gartner, Inc. Implicit profile for use with recommendation engine and/or question router
US9721263B2 (en) * 2012-10-26 2017-08-01 Nbcuniversal Media, Llc Continuously evolving symmetrical object profiles for online advertisement targeting
US9412092B2 (en) 2013-01-14 2016-08-09 Google Inc. Generating a filtered view of a content stream
US20150142559A1 (en) * 2013-11-20 2015-05-21 Snecma Customers comparison and targeting method
US9710546B2 (en) * 2014-03-28 2017-07-18 Microsoft Technology Licensing, Llc Explicit signals personalized search
US20160357820A1 (en) * 2015-06-07 2016-12-08 Apple Inc. Reader application with a personalized feed and method of providing recommendations while maintaining user privacy

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6078916A (en) * 1997-08-01 2000-06-20 Culliss; Gary Method for organizing information
US20050216457A1 (en) * 2004-03-15 2005-09-29 Yahoo! Inc. Systems and methods for collecting user annotations
US20070016553A1 (en) * 2005-06-29 2007-01-18 Microsoft Corporation Sensing, storing, indexing, and retrieving data leveraging measures of user activity, attention, and interest

Family Cites Families (38)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6112186A (en) * 1995-06-30 2000-08-29 Microsoft Corporation Distributed system for facilitating exchange of user information and opinion using automated collaborative filtering
US6981214B1 (en) * 1999-06-07 2005-12-27 Hewlett-Packard Development Company, L.P. Virtual editor and related methods for dynamically generating personalized publications
JP4743740B2 (en) * 1999-07-16 2011-08-10 マイクロソフト インターナショナル ホールディングス ビー.ブイ. To create an automated alternative content recommendation and system
US6981040B1 (en) * 1999-12-28 2005-12-27 Utopy, Inc. Automatic, personalized online information and product services
US7721310B2 (en) * 2000-12-05 2010-05-18 Koninklijke Philips Electronics N.V. Method and apparatus for selective updating of a user profile
US7243105B2 (en) * 2002-12-31 2007-07-10 British Telecommunications Public Limited Company Method and apparatus for automatic updating of user profiles
KR100493902B1 (en) * 2003-08-28 2005-06-10 삼성전자주식회사 Method And System For Recommending Contents
US7526458B2 (en) * 2004-11-04 2009-04-28 Manyworlds, Inc. Adaptive recommendations systems
US7299407B2 (en) * 2004-08-24 2007-11-20 International Business Machines Corporation Marking and annotating electronic documents
US7703030B2 (en) * 2005-01-11 2010-04-20 Trusted Opinion, Inc. Method and system for providing customized recommendations to users
US8060463B1 (en) * 2005-03-30 2011-11-15 Amazon Technologies, Inc. Mining of user event data to identify users with common interests
WO2007002820A3 (en) * 2005-06-28 2007-05-18 Yahoo Inc Search engine with augmented relevance ranking by community participation
EP1963957A4 (en) * 2005-12-19 2009-05-06 Strands Inc User-to-user recommender
US7457772B2 (en) * 2006-01-06 2008-11-25 Tabin Joshua Z Method and apparatus for interactive criteria-based commodity comparisons
US8019777B2 (en) * 2006-03-16 2011-09-13 Nexify, Inc. Digital content personalization method and system
US20080077574A1 (en) * 2006-09-22 2008-03-27 John Nicholas Gross Topic Based Recommender System & Methods
US20080208852A1 (en) * 2007-02-26 2008-08-28 Yahoo! Inc. Editable user interests profile
US9405830B2 (en) * 2007-02-28 2016-08-02 Aol Inc. Personalization techniques using image clouds
US8484149B1 (en) * 2007-03-06 2013-07-09 Patrick Laughlin Kelly User interface for entering and viewing quantitatively weighted factors for decision choices
US7836151B2 (en) * 2007-05-16 2010-11-16 Palo Alto Research Center Incorporated Method and apparatus for filtering virtual content
US20070239610A1 (en) * 2007-05-17 2007-10-11 Floga Holdings, Llc Methods, systems and apparatus for displaying user generated tracking information
US20090163183A1 (en) * 2007-10-04 2009-06-25 O'donoghue Hugh Recommendation generation systems, apparatus and methods
US20090112837A1 (en) * 2007-10-24 2009-04-30 Natwar Modani Proactive Content Dissemination to Users
US9081853B2 (en) * 2008-04-03 2015-07-14 Graham Holdings Company Information display system based on user profile data with assisted and explicit profile modification
US20120030553A1 (en) * 2008-06-13 2012-02-02 Scrible, Inc. Methods and systems for annotating web pages and managing annotations and annotated web pages
US8520000B2 (en) * 2008-07-02 2013-08-27 Icharts, Inc. Creation, sharing and embedding of interactive charts
US8037080B2 (en) * 2008-07-30 2011-10-11 At&T Intellectual Property Ii, Lp Recommender system utilizing collaborative filtering combining explicit and implicit feedback with both neighborhood and latent factor models
US20100106668A1 (en) * 2008-10-17 2010-04-29 Louis Hawthorne System and method for providing community wisdom based on user profile
US20120036523A1 (en) * 2009-01-01 2012-02-09 Orca Interactive Ltd. Adaptive Blending of Recommendation Engines
US9519716B2 (en) * 2009-03-31 2016-12-13 Excalibur Ip, Llc System and method for conducting a profile based search
US8489515B2 (en) * 2009-05-08 2013-07-16 Comcast Interactive Media, LLC. Social network based recommendation method and system
US20110184960A1 (en) * 2009-11-24 2011-07-28 Scrible, Inc. Methods and systems for content recommendation based on electronic document annotation
US20110191311A1 (en) * 2010-02-03 2011-08-04 Gartner, Inc. Bi-model recommendation engine for recommending items and peers
EP2646964A4 (en) * 2010-12-01 2015-06-03 Google Inc Recommendations based on topic clusters
US9292602B2 (en) * 2010-12-14 2016-03-22 Microsoft Technology Licensing, Llc Interactive search results page
US9158775B1 (en) * 2010-12-18 2015-10-13 Google Inc. Scoring stream items in real time
US20130073568A1 (en) * 2011-09-21 2013-03-21 Vladimir Federov Ranking structured objects and actions on a social networking system
US20130166346A1 (en) * 2011-12-22 2013-06-27 Saudi Arabian Oil Company Systems, Computer-Implemented Methods and Computer-Readable Media to Provide Multi-Criteria Decision-Making Model for Outsourcing

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6078916A (en) * 1997-08-01 2000-06-20 Culliss; Gary Method for organizing information
US20050216457A1 (en) * 2004-03-15 2005-09-29 Yahoo! Inc. Systems and methods for collecting user annotations
US20070016553A1 (en) * 2005-06-29 2007-01-18 Microsoft Corporation Sensing, storing, indexing, and retrieving data leveraging measures of user activity, attention, and interest

Also Published As

Publication number Publication date Type
US20140108395A1 (en) 2014-04-17 application

Similar Documents

Publication Publication Date Title
Tene et al. Big data for all: Privacy and user control in the age of analytics
US8355955B1 (en) Method, medium, and system for adjusting a selectable element based on social networking usage
US7555478B2 (en) Search results presented as visually illustrative concepts
US20070124208A1 (en) Method and apparatus for tagging data
US20090089293A1 (en) Selfish data browsing
US20100241507A1 (en) System and method for searching, advertising, producing and displaying geographic territory-specific content in inter-operable co-located user-interface components
US20110295612A1 (en) Method and apparatus for user modelization
US20140089816A1 (en) Displaying social networking system entity information via a timeline interface
US20120191764A1 (en) Systems and methods for managing profiles
US20090077062A1 (en) System and Method of a Knowledge Management and Networking Environment
US20070067331A1 (en) System and method for selecting advertising in a social bookmarking system
US20100082684A1 (en) Method and system for providing personalized web experience
US20110015996A1 (en) Systems and Methods For Providing Keyword Related Search Results in Augmented Content for Text on a Web Page
US20080004884A1 (en) Employment of offline behavior to display online content
US20080005313A1 (en) Using offline activity to enhance online searching
US20120030018A1 (en) Systems And Methods For Managing Electronic Content
US20090254529A1 (en) Systems, methods and computer program products for content management
US20100122174A1 (en) System and method for interfacing interactive systems with social networks and media playback devices
US20100293221A1 (en) Apparatuses, Methods And Systems For Integrated, Information-Engineered And Self-Improving Advertising, E-Commerce And Online Customer Interactions
US20140129328A1 (en) Providing augmented purchase schemes
US20070146812A1 (en) Reader editable advertising
US20110320441A1 (en) Adjusting search results based on user social profiles
US20090307762A1 (en) System and method to create, save, and display web annotations that are selectively shared within specified online communities
US20130097285A1 (en) Media enrichment system and method
US20100058195A1 (en) System And Method For Interfacing A Web Browser Widget With Social Indexing

Legal Events

Date Code Title Description
AS Assignment

Owner name: GARTNER, INC., CONNECTICUT

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:POLONSKY, NIR;MALANGI, GIRISH;VOLPE, KEVIN;SIGNING DATESFROM 20130927 TO 20131001;REEL/FRAME:031348/0804

AS Assignment

Owner name: JPMORGAN CHASE BANK, N.A., AS ADMINISTRATIVE AGENT

Free format text: SECURITY INTEREST;ASSIGNOR:GARTNER, INC.;REEL/FRAME:039041/0250

Effective date: 20160617