Effort based relevance

• US 7,809,733 B2
• Filed: 03/02/2006
• Issued: 10/05/2010
• Est. Priority Date: 03/02/2006
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Lu, Xiaohua, Arun, Gopalan, Mor, David, Carrer, Marco
• Primary Examiner(s): Mahmoudi; Tony
• Assistant Examiner(s): Weinrich; Brian E
• Application Number:
  US11/366,096
• Publication Number:
  US 20070226205A1
• Time in Patent Office: 1,678 Days
• Field of Search: 707/749, 707/767
• US Class Current: 707/749
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

Systems, methods, media, and other embodiments associated with effort based relevance in discussion forums are described. One example method includes receiving activity data concerning messages associated with electronic discussion forum topics. The activity data may include a view count (Vw), a book mark count (Bf), a reference count (Rf), and a reply count (Rp). The effort based relevance may be computed from Vw and one or more of, Bf, Rf, and Rp. Topics may be logically ordered using the effort based relevance.

Claims

1. A method, comprising:

receiving activity data concerning messages associated with members of a set of topics associated with an electronic discussion forum, the activity data including a view count (Vw), a book mark count (Bf), a reference count (Rf), a reply count (Rp), and rating data (Rd);

computing effort based relevance for a configurable period of time for members of the set of topics based on Vw, Bf, Rf, and Rp;

weighting effort based relevance based on Rd;

logically ordering the set of topics with respect to effort based relevance;

acquiring user data concerning a user visiting the discussion forum, the user data including a set of viewed topics, a set of viewed messages, and a set of queries provided by the user;

determining a context in which the user is operating, the context being associated with a task the user is performing, the context being one of, following a topic, participating in a topic, referencing an element in a topic, and performing a combination of following a topic, participating in a topic, and referencing an element in a topic;

providing a recommendation to the user concerning a set of topics, the recommendation being based, at least in part, on the context in which the user is operating and effort based relevance for members of the set of topics; and

automatically generating a set of frequently asked questions based, at least in part, on effort based relevance for members of the set of topics;

where computing effort based relevance includes;

computing an X axis component of an effort based relevance measurement based on Vw and Rp;

computing a Y axis component of the effort based relevance measurement based on Vw, Bf, and Rf;

computing an X axis partitioning measure (α

) based on the average of (Rp/Vw) for the set of topics;

computing a Y axis partitioning measure (β

) based on the average of ((Rf+Bf)/Vw) for set of topics; and

computing the effort based relevance based on one or more of, a Euclidean distance from a first point described by the intersection of an X axis and a Y axis, and a Euclidean distance from a second point described by the intersection of α and

β

.

2. A computer-implemented method, comprising:

receiving activity data concerning items in a set of topics being acted upon in an electronic discussion forum, the activity data comprising a view count (Vw), a book mark count (Bf), a reference count (Rf), and a reply count (Rp), where a reply count corresponds to a number of replies for an item in the electronic discussion forum;

combining the activity data to determine effort based relevance coordinates for each item, the effort based relevance coordinates defining a position of the item in a two-dimensional representation of relevance comprising a two dimensional X-Y graph, where the effort based relevance coordinates are determined by;

computing an X axis component of the effort based relevance coordinates based, at least in part, on Vw and Rp; and

computing a Y axis component of the effort based relevance coordinates based, at least in part, on Vw, Bf, and Rf;

assigning a first effort based relevance to each item based, at least in part on, a distance between the effort based relevance coordinates for the item and an origin of the X-Y graph;

ranking the items according to the first effort based relevance; and

providing the items to a user of the electronic discussion forum in order of ranking.

3. The computer-implemented method of claim 2 comprising:

assigning an item relevance characteristic to each item based, at least in part on, a region of dimensional representation of relevance in which the effort based relevance coordinates for each item are positioned;

determining a user relevance type for a user participating in the electronic discussion forum;

selecting items having an item relevance characteristic that matches the user relevance type; and

providing the selected items to the user in order of ranking.

4. The computer-implemented method of claim 3 where the two-dimensional representation of relevance comprises a two dimensional X-Y graph and further where the item relevance characteristic is assigned based, at least in part, on a quadrant of the X-Y graph in which the item is positioned.

5. The computer-implemented method of claim 2, the activity data including rating data (Rd) and where the method includes weighting the first effort based relevance measurement using Rd.

6. The computer-implemented method of claim 2 further comprising providing a two-dimensional plot of the relative position of the effort based relevance coordinates for the items.

7. A computer-implemented method, comprising:

receiving activity data concerning items in a set of topics being acted upon in an electronic discussion forum, the activity data comprising a view count (Vw), a book mark count (Bf), a reference count (Rf), and a reply count (Rp), where a reply count corresponds to a number of replies for an item in the electronic discussion forum;

combining the activity data to determine effort based relevance coordinates for each item, the effort based relevance coordinates defining a position of the item in a two-dimensional X-Y graph of relevance, where the X-Y graph is divided into quadrants by;

computing an X axis partitioning measure (α

) based, at least in part, on one or more of an average or median of (Rp/Vw) for the set of topics; and

computing a Y axis partitioning measure (β

) based, at least in part, on one or more of an average or median of ((Rf+Bf)/Vw) for the set of topics;

assigning a first effort based relevance to each item based, at least in part on, the position of the item'"'"'s effort based relevance coordinates in the two-dimensional representation of relevance;

assigning an item relevance characteristic to each item based, at least in part on, a quadrant of the X-Y graph in which the effort based relevance coordinates for each item are positioned;

determining a user relevance type for a user participating in the electronic discussion forum;

selecting items having an item relevance characteristic that matches the user relevance type; and

providing the selected items to the user in order of ranking.

8. The computer-implemented method of claim 7 comprising assigning a second effort based relevance to an item based, at least in part, on a distance between the item'"'"'s effort based relevance coordinates and a second point described by the intersection of α

and β

.

9. The computer-implemented method of claim 8, including computing a combined topic relevancy based, at least in part, on the first and second effort based relevance and a keyword based relevance measurement.

10. The computer-implemented method of claim 8 where the user relevance type is determined by acquiring user data concerning a user visiting the discussion forum, the user data including information concerning one or more of, a set of viewed topics, a set of viewed items, and a set of queries provided by the user;

the method further comprising providing a recommendation to the user concerning one or more items, the recommendation being based, at least in part, on the context in which the user is operating and the second effort based relevance for the one or more items.

11. The computer-implemented method of claim 8, including automatically generating a set of frequently asked questions based, at least in part, on the second effort based relevance for items in the set of topics.

12. A computer-readable medium storing processor executable instructions that when executed by a computer cause the computer to perform a method, the method comprising:

receiving activity data concerning items in a set of topics where the activity data comprises a view count (Vw), a book mark count (Bf), a reference count (Rf), and a reply count (Rp), where a reply count corresponds to a number of replies for an item in the electronic discussion forum;

combining the activity data to determine effort based relevance coordinates for each item, the effort based relevance coordinates defining a position of the item in a two-dimensional representation of relevance comprising a two dimensional X-Y graph, where the effort based relevance coordinates are determined by;

computing an X axis component of the effort based relevance coordinates based, at least in part, on Vw and Rp;

computing a Y axis component of the effort based relevance coordinates based, at least in part, on Vw, Bf, and Rf;

assigning a first effort based relevance to each item based, at least in part on, a distance between the effort based relevance coordinates for the item and an origin of the X-Y graph;

ranking the items according to the first effort based relevance;

providing the selected items to the user of the electronic discussion forum in order of ranking.

13. The computer-readable medium of claim 12 where the instructions include:

computing an X axis partitioning measure (α

) based, at least in part, on one or more of an average or median of (Rp/Vw) for the set of topics;

computing a Y axis partitioning measure (β

) based, at least in part, on one or more of an average or median of ((Rf+Bf)/Vw) for the set of topics;

assigning a second effort based relevance to each item based, at least in part, on a distance to a second point described by the intersection of α and

β

; and

computing a combined topic relevancy based, at least in part, on the first and second effort based relevance and a keyword based relevance measurement.

14. The computer-readable medium of claim 12 where the instructions include:

assigning an item relevance characteristic to each item based, at least in part on, a region of dimensional representation of relevance in which the effort based relevance coordinates for each item are positioned;

determining a user relevance type for a user participating in the electronic discussion forum;

selecting items having an item relevance characteristic that matches the user relevance type; and

returning the selected items to the user.

15. A system, comprising:

a means for tracking user actions taken on items in an electronic discussion forum, the actions including a view count (Vw), a book mark count (Bf), a reference count (Rf), and a reply count (Rp), where a reply count corresponds to a number of replies for an item in the electronic discussion forum;

a means for combining (Vw), (Bf), (Rf), and (Rp) to determine effort based relevance coordinates for each item, the effort based relevance coordinates defining a position of the item in a two-dimensional representation of relevance assigning an item relevance to each item based, at least in part on, the position of the item in the two-dimensional representation of relevance comprising a two dimensional X-Y graph, where the effort based relevance coordinates are determined by;

computing an X axis component of the effort based relevance coordinates based, at least in part, on Vw and Rp; and

computing a Y axis component of the effort based relevance coordinates based, at least in part, on Vw, Bf, and Rf;

a means for ranking the items according to the item relevance; and

a means for recommending selected items to a user of the electronic discussion forum in ranking order.

16. A computer-readable medium storing processor executable instructions that when executed by a computer cause the computer to perform a method, the method comprising:

receiving activity data concerning items in a set of topics being acted upon in an electronic discussion forum, the activity data comprising a view count (Vw), a book mark count (Bf), a reference count (Rf), or a reply count (Rp), where a reply count corresponds to a number of replies for an item in the electronic discussion forum;

combining the activity data to determine effort based relevance coordinates for each item, the effort based relevance coordinates defining a position of the item in a two-dimensional X-Y graph of relevance, where the X-Y graph is divided into quadrants by;

computing an X axis partitioning measure (α

) based, at least in part, on one or more of an average or median of (Rp/Vw) for the set of topics; and

computing a Y axis partitioning measure (β

) based, at least in part, on one or more of an average or median of ((Rf+Bf)/Vw) for the set of topics;

assigning a first effort based relevance to each item based, at least in part on, the position of the item'"'"'s effort based relevance coordinates in the two-dimensional representation of relevance;

assigning an item relevance characteristic to each item based, at least in part on, a quadrant of the X-Y graph in which the effort based relevance coordinates for each item are positioned;

ranking the items according to the first effort based relevance;

determining a user relevance type for a user participating in the electronic discussion forum;

selecting items having an item relevance characteristic that matches the user relevance type; and

providing the selected items to the user in order of ranking.

17. The computer-readable medium of claim 16 where the instructions include assigning a second effort based relevance to an item based, at least in part, on a distance between the item'"'"'s effort based relevance coordinates and a second point described by the intersection of α

and β and

computing a combined topic relevancy based, at least in part, on the first and second effort based relevance and a keyword based relevance measurement.

18. The computer-readable medium of claim 16 where the user relevance type is determined by acquiring user data concerning a user visiting the discussion forum, the user data including information concerning one or more of, a set of viewed topics, a set of viewed items, and a set of queries provided by the user;

the instructions further comprising providing a recommendation to the user concerning one or more items, the recommendation being based, at least in part, on the context in which the user is operating and the second effort based relevance for the one or more items.

19. The computer-readable medium of claim 16 where the instructions include automatically generating a set of frequently asked questions based, at least in part, on the second effort based relevance for items in the set of topics.