• \ .
·"EXPRESS MAIL NO. EV528712217US
METHOD AND SYSTEM FOR PRIORITIZING COMMUNICATIONS BASED ON
INTERPERSONAL REUHiONSHIPS
TEC.HNICAL FIELD
[D001.J The ~ascribed technology relates to prioritizing communications, such a·s
electronic mail messages.
BACKGROUND .
{0002J A person can receive many hundreds of electronic· communications each
day. The electronic communications can include electronic mail messages, voice
mail messages, memoranda, docume.nts, and so o·n. · The communications are
typically sent from a sender (e.g., a p~rson, group of persons, or organization) to
one or more recipients (e.g., a person, group of persons, or organization).
.B e~ause of the large number .o f communications, especially electronic mail
messages that a person can receive, it can be very time-consuming for a user to
access and process all their communications. Indeed, because of the large
number of electronic mail messages, it may be difficult for a user to identify an
important electronic mail message that may need prompt attention out of all the
electronic mail. messages of lesser importance. To help locate messages, some
electronic mail systems allow a person to specify the order in which electronic mail
messages are displayed. For example, a person can specify to order electronic
mail messages based on time of delivery, sender, subject, and so on.
[0003} These techniques for ordering electronic mail messages and other
communications do not, howeve_r, provide much useful insight into the importance
of cqmmunications. For example, if an employee wants to view electronic mail
_ messages sent from their supervisor as soon as possibJe, the employee may need
to order the inbox based on sender and. then review the list of messages to see if
any were sent by the supervisor. It would be desirable to provide a technique that
141826-8054 USOO/SL041910.028f" 7/30104
would allow for communications to be automatically prioritized so a person ean
focus their attention on communications that are important before focusing their
attentic:>n on communications of lesser importance.
SUMMARY
[0004) A method and system for cal~ulating· ·the importance of persons based on
interpersonal relationships is provided. The interpersonal relationships may
include participant relationships, distribution relationships, and organizational
relationships as described below. The system may represent interpersonal
relationships as links between persons and apply a link-based ranking algorithm to
calculate ~he .importance of the persons. When a person receives a
communication, the system can prioritize the communication relative to· other
communications based on the importance of the participants of the communication
such as the sender.
BRIEF DESCRIPTION OF THE DRAWINGS
rooos) Figure 1 is a block diagram that illustrates components of the prioritization
system in one embodiment.
· rooos] · Figure 2 is a flow diagram that illustrates the prioritize electronic mail
messages component in one embodiment.
(00071 Figure 3 is a flow diagram that illustrates the processing· of a calculate
importance based on recipient relationship component in one embodiment.
roooBJ Figure 4 is a flow diagram that- illustrates·the processing of the generate
recipient matrix component in one embodiment.
[0009J Figure 5 is a flow diagram that illustrates the processing of the calculate
importance based on distribution relationship component in one embodiment.
[0010J Figure 6 is a flow diagram that illustrates the processing of the generate
distribution matrix component in one. embodiment.
f0011J Figure 7 is a flow diagram that illustrates the processing of the calculate
· importance based on organiz~tional relationship component in one embodiment.
!41826-8054 USOO/Sl041910 028) -2- 7130/04
···--·----~~~~~------------~-------------
.,
DETAILED DESCRIPTION
{0012)_ A method and system for calculating the importance of persons based ori
interpersonal relationships and prioritizing communications based on importance
of participants in the communications is provided. In one embodiment, a
prioritization system identifies relatio~ships between persons and identifies the
. importance .. of a person to other persons based on these relationships. A . .. ..
relations~ip between two persons may be that one person ~s a recipient of a
communication sent py another person who is a sender.. Thus, the recipient has a
"recipient" .relationship to the sender. A person who has a recipient relationship
with many senders on many communications may be considered an "important"
person. Thus; it may be ~esirable_ to promptly review a communic~tion from such
an important person. ·Also, a person who receives communications from other
important persons may themselves be important. Importance could also be based
on a "sender" relationship in that a p.e rson who sends a lot of communications to .
_other persons, especially important persons, may be important. The sender and
recipient relationships are referred to as "participant" relationships. Another
re.lationship between two _persons may be that both of them are members of the
same distribution list such as an electr_onic mail distribution list. Thus, the persons
have a "distribution" relationship to each other. A person who has a distribution
relationship with many other persons on many distribution lists, especially other
important persons, may be considered to be an important person. Another
relationship between two persons may be that they are members of the same
organization that can be represented by an organizational chart.. Thus, two
persons in the same organization have an "organizational" relationship. The
importance of one person to another person within an organization may be based
on the distance between the persons within the organizational chart hierarchy.
After the prioritization system identifies the importance of persons, the prioritization
system can prioritize communications based on the importance of the senders or
recipients. . The prioritization system may set the priority of a communication
based on the importanc~ of the source (e.g., sender, originator, creator) of the
communication.. The ~ource ·may_ be a person other than the sender. For
(41826-8054 USOO/Sl041910.028J . -3- 7130104
••
-example, an executive assistant may. se'"'d an electronic mail message. on b~half
of an executive who cr:eated the message. If multiple persons are associated with
the source (e.g., an electronic mail message sent from a group), the prioritization
system may set the priority based on an aggreg~te importance of the members of
the group. The prioritization system may also base the priority of a communication
on_ the importance of the _targets (~.g., recipient, re~ipient's supervisor) ~f the
communication. For exampl~. se~ders may send electronic. mail -messages
intended for an executive to the executive's assistant. If a communication is sent
to many important persons, then the communication is more likely to be of high
interest to a recipient based on the aggregate importance of the recipients. In the
following, the prioritization system is described in th.e eontext of an electronic m~il
· system. One skilled in the art will appreGjate, however, that the prioritization
system-can be used in the context of other communication systems.
[00131 . In one embodiment, the prioritization ~ystem calculates the importance of
persons by applying a ranking algorithm to participant relationships and in
particular to recipi~nt relationships. The prioritization system may generate a
matrix with rows and columns representing persons with each element at the
intersection of a row and column representing the number of times that the person
of the column is a recipient of an electronic mail message in which the person of
the row is a sender. The prioritization system may generate the matrix ·based on
analyzing electronic mail messages of all the persons within an organizatio"n.
When privacy,· confidentiality, or other concerns do not allow access to such .
e)ectronic mail messages, the prioritization system can generate the matrix based
on electronic mail messages sent or received by a single person. Also, the
· prioritization system may aggregate matrices that are each based on the electronic
mail messages of a single person into an aggregate matrix for an organization or
portion of an organization whose members agree to participate in the aggregation.
The prioritization system can co!lect the matrices in a way that helps ensure the
privacy of the individuals. After the matrix is generated, the prioritization s~stem
applies a ranking algorithm to the matrix to calculate . the importance of each
person. The ranking algorithm may be a linked-based ranking algorithm. such as a
(41 826-8054 US001SL041910 026] ~.._:,. -4- 7130/04
PageRank-type algorithm or a HITS-type algorithm applied to the interpersonal
. .
relationships. rather than to link relationships .o f. web pages. The. interpersonal
relationships. are represented as links between· persons. The· PageR~mk and HITS
algorithms are described below. -·
[0014J In another embodiment, the prioritization system calculates the importance
of. persons by applying a ranking algorithm to the distribution rel~tionships. The
prioritization· system ~ay generate a matrix with row~ and columns representing
persons with each element at the intersection of a row·and Cc;>lumn representing
the number of times that the person of the column is. on the same distribution li.st .
. as the person _of the.row. The prioritization system may generate the matrix based
on analyzing electronic mail distribution lists of an organizati~n. After the matrix is
generated, the prioritization system applies· a ranking algorithm to the matrix to
calculate the importance of each person. The ranking algorithm may be a linkedbase_
d ranking algorithm such as a PageRank-type algorithm or a HITS-type
algorithm applied to the interpersonal relationships rather than to link relationships
of web pages.
. .
{0015J In another embodiment, the prioritization system calculates the importance
of one person to another person based on organizational relationships. The
prioritization system may use an electronic representation of an organizational
. .
· chart to identify the relationship between two persons. The organization
relationship may be established when the persons are in the same organization,
and a reporting relationship may be established when one · person of the
organization reports to another person of the organization directly or indirectly.
For example, an employee and the employee's supervisor may have a reporting
relationship. The importance of one person who has an organizational relationship
to another person may be based on the difference in their levels within .tt)e
hierarchy of the organization and based on how many persons are at the same
· level. For example, a supervisor of an employee may be important to the
employee because the employee has only one supervisor. However, the ·
employee may be less important (in terms of communications) to the supervispr
because the supervisor may supervise many employees and each ·supervised
141826:8054 USOOISL041910.028} -5- 7130/04
. .
employee may have the same importance to the supervisor. As another. example,
a supervi~or of an employee may be ~ore importa.nt to the employee t.han the
supervisor's. supervisor because the employee reports only indirectly to the
supervisor's supervisor. The importance based on an organizational relationship
may be represente~ by the following equation: ·.
(1)
where a1(i) represents the importance of person ; . to person j I len(i,j)
represents · the distance or length from person ; to person j I and
· j{kyen(k.J) = len(i.J)}j is the number of persons the same ~istance and .directio.n
away from person} as person;. For example, the distance between a superyisor
and an employee i.s 1, and the dista~ce be~een the sup~rvisor's supervisor and
the employee is 2. Thus, the importance of the supervisor to the employee .is 1,
· · · but the· importance of the employee to the ~upervisor who supervises 5 employees
is 1/5. Further, the importance of the supervisor's supervisor to the employee· is
1/2, and the importance of the employee to the supervisor's supervisor is 1120,
when the supervisor's supervisor has 1 0 employees at the same level of the
organization chart.as the employ.ee. One skilled in the art will appreciate that the
importance based on an organizational relationship can be defined in many
different ways. For exa~ple, the importanc~ can decrease exponentially based on
distance within the hierarchy between two persons. The distance may also be
limited to a reporting distance b~tween persons with a reporting relationship. For
example, two employees who report to the same supervisor would not have a
reporting relationship and thus the importance based on the reporting relationship
would be 0. However, if a non-reporting relationship is used, then the distance
betw~en them would be 2 (i.e., 1 from an employee to a common supervisor and 1
from the common supervisor to the other employee), and their importance would
be 1/10, when there are 5 employees at the same level.
(001~] In one embodiment, the ~mportance of a. person can be based on a
combination of various methods for calculating importance. For example. the
141826-8054 USOOJSL041910.028] -6- 7130/04
prioritization system could· calculate the importance of a . person by taking a
weighted average of th.e · importances based · on participant relationships,
distribution relationships, organizatio·nal relationships, and so on. The weight
applied to, each importance may reflect the confidence that it accurately reflects
the real importance of a person. For example, if a· participant relationship is
considered twice as accurate as a dis~ribution relationship or an organizational
relationship, then the weights for the participant, dist~~ution, and organizational
relationships may be .5, .25, and .25. The importances may also be normalized to
a value between 9 and 1 to facilitate 'their combining. The weights can be
identified by a regression method based on training data. Regression tries to
·determines the relation~hip be~eel"! two ra~dom variab!es x = (x1 ,~2 .... xP) andy.
. .
A linear regression method explains the relationship between x ~nd y ·with a
. .
straight line fit to the .- training data. The linear regression meth9d postulates that: · .
p
y = h0 + L h1x1 + e
J~l
(2)
where the "residual" e is a random variable with a mean of zero and the
coefficients bi ( o ~ J ~ p) are determined by the condition that the sum of. the
square of the residuals is as small as .possible. Therefore, the linear combination
with b, should be better than those with any other coefficients. The variable x can
come directly from inputs, or son:te transformations of inputs, such as a logarithmic
or a polynomial transformation.
(0017J Figure 1 is a block diagram that illustrates components of the prioritization
system in one· embodiment. The prioritization system 100 includes a prioritize
electronic mail messages component 101, a calculate importance based on
participant relationship component 102, a calculate importance based on
distribution relationship component 103, and a calculate importance based· on
organizational relationship component 1 04. The prioritize electronic mail
messages ~omponent may be invoked periodically to prioritize the messages in an
electronic mail in box 110. The component may prioritize the messages in the
inbox based on the information of an !mportance store 111 and prioritize each
j41826·8054 USOO/SL041910 028) -7- 7/30/04
message as it is received. The component may pri~ritize the messages based on
the importance of the sender as indicated by the importance store. The
importance information of the importance store may be .calculated by one of the
calculate importance components. Alternatively, the importance of the persori
may·· be ·based on a combination· of the importances calculated by different
calculate importance components. The calculate importance based on participant
relationship component may generate a matrix of the .sender-recipient
relation~hips based on the· inform~tion of an electronic mail message store 112.
Tl:le electronic mail message store may contain all electronic ma!l messages sent
or receive~ by an o~g~~i-~ation or an indivi~ual. The calculate importance based
on participant relation·ship component then applies a ranking algorithm to calculate
an importance for each person. The calculate importance based on distribution
relationship component may generate a matrix of the distribution relationships
based on information in a distribution· list store 113. The calculate importance
based on distribution _relationship component may generate a matrix based on
. persons being on the same distribution list. The calculate importance based on . .
distribution relationship component then applies a ranking algorithm to calculate
the importance of each person. The calculate importance based on organi~ational
relationship component calculates the importance of persons based on the
information of an organizational chart store 114.
(0018J The. computing device on which ·the prioritization system is implemented
may include a central processing unit, memory, input devices (e.g., keyboard and
pointing devices), output devices (~.g., display devices), and storage devices (e.g.,·
disk drives). The memory and storage devices are computer-readable media that
..
. may contain instructions that implement the prioritization system. In add!tion, the
-
data structures and message structures may be stored or transmitted via a data
trans!T1ission medium, such as a signal on a communications link. Various
communications links _may be used, such as the Internet, a local area network, a
wide area network, or a point-to-point dial-up connection.
[0019J The prioritization system may be implemented in various operating
environments t_hat include_ personal computers, server computers. hand-held or
(41826-8054 USOOiSL04191CJ.028) -8- 7/30104
..
. (0020]
laptop devices, multiprocessor systems, microprocessor~based ·.systems,
programmable consumer electronics, network PCs, minicompute-rs, mainframe_
computers, distributed computing environments that include any of the above
systems or devices, and the like.
The prioritization system may: be descri~ in the general context of
"computer-executable instructions, such as progra~ modules,· executed by one or
more computers or other devices. Generally, program modules include. routines,
programs, objects, components, data structures, and so on that perform particul~r
tasks or implement particular abstract data types. Typically, the-functionality of ttie
program modules may be combined or distributed as ·desired in various
embodiments.
(00211 Two well~known techniques for ranking web pages are PageRank and .HITS
("Hyperlink~fnduced Topic Search").· The prioritization system may apply th~se
. .
algorithms to the participant and distribution relationsh.i p matrices to calculate ~he .
importance or rank of the persons b~sed on the relationship. PageRank is based
on the principle that web pages will have links {i.e., "outgoing links") to important
web pages. Thus, the importance of a web page is based on the number and
importance of other web pages that link to that web page (i.e., "incoming links").
Similarly, the importance of a person can be based on the number· of and
importance of other persons who send electronic mail messages to that person.
Thus, the web pages.of these ranking algorithms can be replaced by persons and
the lin_ks by their interpersonal relationship. In a simple forr:n, the links between
w~b pages can be represented by matrix A I where A,1 represents the number of
outgoing links from web page i to web page j. The importance score w
1
for web
page J can be r~presented by the following equa_tion:
w=LAw I I ~ I
This equation can be solved by iterative calculations based on the. following
equation:
(41826-6054 USOO/SL041910 02BJ -9- 7130104
where w.is the vector of importance scores for the web pages and !s the principal.
eigenvector of AT. To ensure the iteration will converge, "random walk" is added
when calculating the page score wi.
[0022J The HITS. technique is additionally ·based on thf;'! principle that a web page
that has many links to other important web pages may itself. be important. Thus,
HITS di~ides "importa.n ce" of web pages into tw. o relateq attributes: "hub" .a nd
"authority." ·"Hub" is measured by the "authority" score of the web pages that a.
web page links to, and "authority" is measured by the "hub" score of the web
pages that iink to the wet? page. In· contrast to Page Rank, which calculates the
importance of web· pages independently from the query, HITS ·calculates
importance based on the web pages of the result and web pages that are related
to the web pag~s of the result by following incoming and outgoing links. HITS.
submits a query to a search engine service and uses~the web pages of the results
as the initial set of web pages. HITS adds to the set those web pages that are the
destinations of incoming links and those web ·pages that are the sources of
outgoing links of the web pages of the result. HITS then calculates the authority
and hub score of each web page using an iterative algorithm. The authority and
hub scores can be represented by the following equations:
a(p) == r h(q) and h(p) = ra(q)
q-+p p-+q
where a(!') represents the authority score fpr w~b page p and l1(p) represents
the hub score for web page p. HITS uses an adjacency matrix A to represent
the links. The adjacency matrix is represented by the following equation:
b = {1 if~age i has a link to page},
9 0 otherwise ·
.The vectors a and h correspond to the authority and f:lub ·scores, respectively, of
all web pages in the set and can be represented by' the following equations:
a= AT h and h = A a
Thus, a and h are eigenvectors of matrices AT A and AAT.
[0023] Figure 2 is a flow diagram t~at illustrates the prioritize electronic mail
messages component in one embodiment. The cor:nponent loo.ps selecting each
(41826-8054 USOO/SL041910.028J -10- 7.130/04
mes. sage of an inbox and assigning a priority based ·on the importance· of the . ..
sender as i~dicated by th~ importance store. In block 201. the component selects
the next electronic mail· message. In decision block 202, if all the electronic mail
messages have already been selected, then the component completes, else the
· component continues at block 203. In block 203, the compon~nt retrieves the
importance of the sender from the importance store. ·In block 204, the component
sets the priority of the selected electronic mail message based on the importance
of the sender. The comp. onent may also .b ase the priority of the message ba.s ed . . .
on the importance of other recipients ofthe electronic mail message and based on
whether those recipients are to-recipients or cc-recipients. The component then
loops to block 201 to select the next electronic mail message.
. .
(00241 Figure 3 is a flow diagram that .illustrates the processing of a calculate.
importance based on recipient relationship component in one embodiment. This
component calculates importance based on a participant being a recipient. In
block 301, the component invokes a generate· recipient ·matrix component to
generate a matrix that indicates the recipient relationships between senders and
recipients of electronic mail messages in the electronic mail message store. In
block 302, .the component applies a ranking algorithm to the matrix to calculate the
importance of each person. The component then completes.
(0025J Figure 4 is a flow diagram that illustrates the processing of the generate
recipient matrix component in one embodiment. The component generates a
square matrix of per:sons that are senders and recipients of the electronic mail
. .
messages of the electronic mail message store. In block 401, the component
generates a square matrix of the electronic mail message participants (i.e.,
senders and recipients). In blocks 402-406, the component loops selecting each
electronic mail m~ssage and updating the matrix accordingly. In block 402, the
component selects the next electronic mail ~essage. In d~cision block 403, if all
the electronic mail messages have already been selected, then th_e ·component
.· .
returns, else the component continues at block 404. In block 404, the·component
selects_ the next recipient of the selected electronic mail message.' In decision
block 405, if all _the recipients have already .been selected, then the component
(41826-8054 USOOfSL04\910 028} 7130104
loops to block 402 to select the next recipient, else the. comp9nent continues at
block 406.. In block 406, .the component increments the. value of the. -matrix
indexed by the sender and the selected recipient of the selected electronic mail•
message. ·The component then loops to block 404 to select the next recipient. In
one embodiment, if there are multiple senders, the component updates the
element for each sender and recipient combination of the matrix. The component
may also increase the value less when the recipient is a cc-recipient rather than a
to,.recipient
.(0026J Figure 5 is a flow diagram that illustrates th~ proces~ing of the calculate
-~ .
importance based on distr.ibution relati~nship component in one embodiment. In
block 501, the component invokes a· generate distribution matrix component to · ..
generate a ~atrix· that indicates the distribution relationships between persons on
the same distribution list. In block 502, the component applies a ranking algorithryl
to the matrix to calculate the importance of each person based on th~ir distribution
relationships. The component then completes.
(0027) Figure 6 is a flow diagram that illu~trates the processing of the generate
distribution matrix component in one embodiment. The component generates a
square matrix of persons that are on a distribution list within the distribution list
store. In block 601, the component creates a square matrix of the persons on the
distribution lists. In blocks 602-608, the component loops selecting each
distribution list and updating the matrix accordingly. In block 602, the component
selects the next distribution list of the distribution list store. In decision block 603,
if all the distributio!llists have already been selected, then the component returns,
el~e the component continues at block 604. In blocks 604-608, the component
loops selecting each person on the selected distribution list and updating the
matrix element for each other person on the selected distribution list. In block 604,
the component $elects the next person on the selected distribution list. In decisi~n
blo"ck 605, if all the persons have already been selected, then the component
loops to block 602 to select the next distribution list, else th·~ component continues
at block 606. In block 606, the component chooses the next person on the
distribution list other then the selected person. In decision block 607, if all the
(41 826-8054 USOO/SL041 910 028J -12- 7130104
•
- -
persons· ·other than the ~elected person have already been chosen, then the
· component loops to block 604 · to. select the next pe.rson, else .the component
continues at block 608. In block 608, the component· inerements the value of the
element of the matrix ir:tdexed by the selected person and the chosen person to
indicate the dist~ibution relationship. The component then loops· to block 606 to
choose the next person of the selected distributiOIJ list. The component may
increase the value of the element of the matrix according to the number of persons
on the distribution list. For example, if the distribution list contains only two
persons, then each person may ·be niore important to the other than -if :the
distribution list contains a thousand persons.
. .
(0028) Figure 7 is a flow diagram that illustrates the processing of the calculate
importance based on organizational relationship component in one embodiment.
The component loops selecting each pair of persons of the . organization,
calculating their distance and numb~r of persons at a similar level, .and calculating
the importance based on the distance and number. In· block 701, the co~ponent
cr~ates a square matrix of the persons. In block 702, the component selects the
next person in the organization. In decision block 703, if all the persons in the
organization have already been selected, then the component returns, else the
component continues at block 704. In block 704, the component chooses the next
person of the organization for the selected person. In decision block 705, if all.the
persons have already been chosen, then the component loops to block 702 to
. . .
select the next person, else the component continues at block 706. In block 706,
the component calculates the distance or length between the selected person and
the chosen person. The distance may be defined. as the sum of the levels
between the selected person and the chosen person and their cro·sest common
ancestor {referring to parent and child relationships of a tree representing the
organizational hierarchy). For example, if the selecte~ person is 2 levels down
and the chosen person is 1 level down from their closest common ancestor, then
their distance wo.uld be 3. If the selected person is an ancestor of the chosen
·per~on, then the selected person would be considered the ·closest. common
ancestor. Fo'r example, if the chosen person is 2 levels down from the selected
(418?6-8054 USOO/Sl041910 028) 7130104
person, then the selecte~ person would be 0 levels down and_ the chosen person
would be 2 levels down, giving a distance of 2. In· block 707: the compo_n~nt
calculates the number of persons considered at the same level fqr purposes of"
Eq. ua.tion 1 for the selected and chosen per~ons. The number may be defined as ~
the sum of the number of persons at the same level down from their closest
co·mmon ancestor. ·For ex_~mple, if the selected person is ·2 levels· down and th~
chosen person is 1 level down from their closest common ancestor .~nd _the
selected person has 25 persons at the same level and the chosen person has 7
persons at the same level, then the number would be 32. If the selected person is
an ancestor of the chosen person, then the number may be set to 1. In bl~ck 708,
the component calculates the importance ba~ed on Equation 1 and then loops to
block 704 to choose the next person.
(00291 One skilled in the art will appreciate that although specific embodiments of
the prioritization system have been described herein for purposes of ilfustfa~
· various modifications may be made without deviating from t~ .~pirit and scC?pe of )
the invention. Accordingly, the invention is not limited except by the- appended
claims.
(41826-8054 USOOtSL041910 028J -14- 7130104

[c1]
[c2)
[c3)
[c4)
[c5)
[c6)
[c7J
CLAIMS
1/We claim:
1. A method in -a computer system for determining priority of a
communication, the method comprising: ·
identifying relationships between people; and
calculating the priority of the communication based on the identified
relationships of a participant in the commllnication.
2 .. The method,,of ~l~irrlj 1 wherein· an identified relationship_ is. a
participa_nt relationship.
3. rhe method of claim ~ wherein the participant relationship is based
on an ~Lectronic mail message: · . •l J•'; ( • ' I f\"• •
\ /!o \\, , - ~ I J!'l - · I ,
A .t '1\ \( ;,
('-' I ' ii
4. The method of claim 2 wherein the calculating of the priority is based
on frequency at which a sender of the communication is a recipient of
communications.
5. The method of claim 4 wherein the calculating of the priority includes
counting nt:Jmber of times a person is a- recipient of a communication.
r ttc\'1--
-, !! I . ;. , .
. ,
-·
' ' c
6. The method of claim 4 wherein the calculating of the priority includes ' I
generating a recipient link matrix representing links from a sender to recipients of
------·-~--- ·- ------ . .
communications and performing a link-based ranking.
7. The method of claim 6 where!n the ranking algorithm is based on a
page ranking algorithm.
1
[41 826-8054 USOO/Sl04191 0.028). · -15-
· (cB] ,8. The method of claim 1 wherein an identified relatjons~ip is a
distribution rel~tionship. ·.•
. [c9) 9. The method of claim 8 wher~in a distribution relationship is based on
electronic mail distribution lists.
(c10) 10. The method of claim 8 wherein the calculating of the priority is based
on number of distribution lists that the sender of the communication is a member
of and number of members of those distribution lists. ·
fc11J 11. The method of claim 8 wherein the calculating of the priority includes
-· .
generating a distribution list matrix representing links between members of the
same distribution list a·nd performing a link-based ranking.
(c12J 12. The method of claim 11 wherein the link-based ranking is based on a .
page ranking algorithm .
. . I
(c13J 13. The method of claim 1 wherein an identified· relationship is an
organizational relationship.
(c14J 14. The method of claim 1 wherein an identified relationship is based on
reporting relationship between a sender of the communication and a recipient of
the communication.
·(c15J 15. The method of claim ·14 wherein the calculating of the priority is
based on distance between the sender and the recipient within the organization.
. •'
[c16J 16. The method o~ claim 15 wherein the calc.ulating of the priority is
further based on number of people with the same distance from the recipient.
(41 626-8054 USOO/SL041910.028( -16- 7130104
·n \\,
·~/
-. [c17) 17. The ·method of claim 14 wherein the calculating of the priority _is
based on importance of a participant defined as
a, (i): l~t~(i,j) ·l{k Jlm:k. j) = let~(i, j)}!
where a1 (i) represents the importance of person i :to person j and /en(i,J)
_represents the reporting distance from person i to-person J.
-------------·--- _,____ -
(c18J 1 B. A ~puier-r~ad~bl~ ~:ediu~',- ~taining in~tru~t~ for controlling a
[c19)
computer system to set a priority of an electronic mail message, by· a· method
comprising:
providing el~ctr~nic mail messages having senders and recipients;
calculating an importance of a sender of the electronic mail message based
on number of the provided electronic mail messages for which the
sender was a recipient; and
establishing the priority of the electronic mail mess~ge based on the
importance_ of the sender.
19. The computer-readable medium of claim 18 wherein the provided
electronic mail messages include electronic mail messages received by a single
person.
(c20J 20. The computer~readable medium of claim 1a wherein the provided
electronic mail messages include electronic mail messages received by more than
one person. ·
(c21J 21. The computer-readable m~dium of claim 18 wherein the calculating
of the import~nce includes generating a recipient link matrix representing links
from a sender to recipients of an electronic; mail message and performir:tg a linkbased
ranking.
(41626-8054 USOO/Sl041910.028J . -17- 7130104
(c22J 22. The compute~-readable medium of claim 18 wherein ·the link-based
[c23)
ranking is page ranking-based.
23. A co~puter-readable medium containing instructions for controlling a
computer system to set a priority of an electronic mail message, by a method
comprising:
providing distribution lists having members; · ..
calculating an importance of a sender of the electronic mail m~ssage based
on number of the provided distribution lists for which the sender is a
member; and
establishing the priority of the electronic mail message based on the
importance of the sender.
[c24J 24. The computer-readable medium of claim 23 wherein the distribution
list is an electronic mail distribution Jist.
fc25J 25. The computer-readable medium of claim 23 wherein the calculating
of importance is based on number of members of the distribution lists of which the
sender is a member.
[~26J 26. The computer-readable medium of claim 23 wherein the calculating
of an importance includes generating a distribution list matrix representing links
between members o.f the same distribution list and performing a link-based
ranking.
[c27J 27. The computer-readable m~dium.of claim 26 wherein the link-based
ranking is page ranking-based.
../
(41 826-8054 USOOIS L041 91 0 02BJ -18- 7130104
[c2BJ 28. A computer-readable mediu~ containing instructions for CC?ntrolling a
computer system to set-~ priority of an-electronic m~il message, by a method
comprising: · .. ·-- . · .
providi~g//"organizati~ information of ari organization indicating
· r~iu11sllip-soeiween members of the organization;
. .
calculating an imp. ortance of a sender of the electronic mail message based.
on a relationship between a sender and a recipient.of the electronic
mail message as indicated by the organization information; and
establishing the priority of the ~lectronic mail message based on the
impo~ance of the sender.
· (c29J 29. The computer-readable medium of claim 28 wherein the relationship
is based on relative positions of members of the organization.
[c30J 30. The computer-readable medium of claim 29 wherein the calculating
of importance is based on a distance between the sender and the recipient within
the organization.
[c31J 31. The computer-readable medium of claim 30 wherein the .calculating
of importance is further based on number of people with the same distance froni
the· recipient.
· [c32]
)
32. The computer-readable medium of claim 30 wherein the calculating-+--· ·);
of the importance is based on · · 1 c.. .. :
a, (i) = !en(i,j) ·/{k ~en~ k, j) = len(i,j)}/
~~" c'u , t· . ' , I(~~ ~'J~eJJii!il'~
where a1 (i) represents the importance of m~mber ; to member j and len(i,j)
represents the reporting dist~nce from member. i to member j .
(41826-6054 USOO/SL041910.028J -19- 7130/04
\ ,..
I I
(c3~f ... ~-. 1 33. A computer system for prioritizing communications, comprising:·
.,' .
' . .
. a component that identifies interpersC?nal relationships;
. .
a component that calculates importance of ·persons b~sed on identified
interpersona~ communications; and
. a c_9mponent that orders communications based on importance of persons
.participating in the communication.
~- . , ·fc34} 34. The computer system of claim 33 wherein a relationship is a
' \ participant relationship.
"'i. ~
.\ \ ·. [c35}
. \~
35. The computer system of claim 34 wherein the participant relationship
.. \,5 is represented as links between persons and the ~mponent .that calculates
importance applies a link-based ranking algorithm.
(c36J 36. The computer system of claim 33 wherein a relationship is a
distribution relationship.
(c37} 37. The computer. system of claim · 36 wherein the distribution
relationship· is represented as links between persons and the component that
. calculates importance applies a link-based ranking algorithm.
[c3BJ 38. The computer system of claim 33 wherein a relationship is an
organizational relationship.
[c39J 39. A computer-readable medium containing instructions for controlling a
computer system to calculate importance of a person by a method of:
identifying interpersonal relationships;
representing the identified interpersonal relationships as links between
persons; and
applying a link-based ranking algorithm to the links to calculate the
importance of a person.
{41826-8054 USOO/SL041910.028J -20- 7130/04
•
. [c40) 40. The computer-readable medium of claim ·39 ' .wherein the
interpersonal relationship is a participant relationship~
.
[c41) 41. The computer-readable . medium of claim 39 . wherein the
interpersonal relationship is. a distribution relationsh. ip.