U.S. patent application number 14/506835 was filed with the patent office on 2015-01-22 for system and method for providing automated clothing fashion recommendations.
The applicant listed for this patent is Atelier CAAAW, Inc.. Invention is credited to Cindy Guo, Jiajing Peng, Annie Wang, Yibo Zhu, Wencen Zou.
Application Number | 20150026084 14/506835 |
Document ID | / |
Family ID | 52111657 |
Filed Date | 2015-01-22 |
United States Patent
Application |
20150026084 |
Kind Code |
A1 |
Guo; Cindy ; et al. |
January 22, 2015 |
SYSTEM AND METHOD FOR PROVIDING AUTOMATED CLOTHING FASHION
RECOMMENDATIONS
Abstract
A client-server social-network for providing automated clothing
suggestions. Fashion interested users form social networks, upload
records of their respective wardrobes and fashion preferences, and
display their wardrobe items and outfits other members of the
network, often displaying entire outfits using customizable virtual
mannequins. The social network members can evaluate the fashion
merits of both their outfits and their friend's outfits, and the
system will further store data pertaining to the social network
group's fashion assessments. The system may also make statistical
inferences as to what types of clothing may be favored and
disfavored by the user's social network group, and present these
recommendations to the user. Other factors, such as weather, event
type, and user's recent history of wearing various wardrobe items
can also be considered. The system can additionally assist in
shopping and gift giving, provide fashion related games, spot
fashion trends, and provide advanced data for fashion
suppliers.
Inventors: |
Guo; Cindy; (Palo Alto,
CA) ; Peng; Jiajing; (Palo Alto, CA) ; Zhu;
Yibo; (Alamo, CA) ; Wang; Annie; (Los Altos,
CA) ; Zou; Wencen; (Los Altos, CA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Atelier CAAAW, Inc. |
Palo Alto |
CA |
US |
|
|
Family ID: |
52111657 |
Appl. No.: |
14/506835 |
Filed: |
October 6, 2014 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
14476984 |
Sep 4, 2014 |
|
|
|
14506835 |
|
|
|
|
Current U.S.
Class: |
705/319 |
Current CPC
Class: |
G06Q 50/01 20130101;
H04N 1/00185 20130101; H04N 1/00244 20130101; G06T 11/60 20130101;
G06Q 30/0631 20130101 |
Class at
Publication: |
705/319 |
International
Class: |
G06Q 50/00 20060101
G06Q050/00; G06T 11/60 20060101 G06T011/60 |
Claims
1. A client-server social-network method for providing automated
clothing suggestions, said method comprising: using at least one
server comprising at least one processor, memory, fashion social
network software, and a network connection to the internet to
provide a fashion social network comprising a plurality of users,
and a database configured to store, on a per user basis, user
clothing information, user fashion preferences, user social network
linkages, and social network fashion preferences; said user
clothing information comprising user wardrobe items, each user
wardrobe item further comprising item type, item color, item style,
item image, and item history information; said at least one server
further configured to receive information pertaining to said user
clothing information, user fashion preferences, and user friend
information from at least one client computerized device, and store
said user clothing information and user fashion preferences and
user friend information in said database; using said at least one
server and said user friend information to establish social network
linkages between said plurality of users, thereby establishing user
social networks; further receiving, on a per user basis, social
network fashion preferences from said user social network, said
social network fashion preferences comprising preferences of at
least one different social network linked contact pertaining to at
least one user wardrobe item, or at least one outfit comprising a
plurality user wardrobe items; wherein said at least one server
uses said user clothing information, user fashion preferences, said
social network fashion preferences and a scoring algorithm to
suggest at least one user wardrobe item, or at least one outfit,
for said user to wear; wherein said scoring algorithm gives
weighting to user wardrobe items or outfits according to a function
of said social network fashion preferences, said user fashion
preferences, and said item history information; further receiving
information pertaining to an actual choice of said user, at a given
time, of at least one user wardrobe item or at least one outfit and
storing said actual choice and time in said database, and using
said actual choice as at least part of said user fashion
preferences.
2. The method of claim 1, further sorting said user wardrobe items
or said outfits according to any of said item types, item colors,
item styles, item history information, user fashion preferences,
and social network fashion preferences, producing sorted user
wardrobe items or sorted user outfits, and using said item images
to display said sorted user wardrobe items or said sorted user
outfits on said at least one client computerized device.
3. The method of claim 2 further displaying said sorted user
wardrobe items or said sorted user outfits on a virtual mannequin
using least one client computerized device; or further uploading
images of said user to said at least one server, and using said
images of said user to customize an appearance of said virtual
mannequin to resemble said images of said user, thereby producing a
customized virtual mannequin, and further displaying said sorted
user wardrobe items or said sorted user outfits on said customized
virtual mannequin using said least one client computerized
device.
4. The method of claim 3, further uploading background images or a
link to a dynamic background for said mannequin to said at least
one server, and further displaying said sorted user wardrobe items
or said sorted user outfits on said customized virtual mannequin
against said background images or dynamic background.
5. The method of claim 1, further receiving a present or reported
future locations of said user and at least some individuals in at
least said user's social network; determining if said present or
reported future locations of said user and at least some
individuals in at least said user's social network are within a
given location criteria; and if said present or reported future
locations of said user and at least some individuals in at least
said user's social network are within said given location matching
criteria, performing any of: a) receiving information pertaining to
wardrobe items or outfits that said user and at least some
individuals in said at least said user's social network are wearing
or plan to wear, and providing a collision warning if at least some
of said item styles or item colors are either similar or identical;
b) receiving information pertaining to wardrobe items or outfits
that said user and at least some individuals in said at least said
user's social network are wearing or plan to wear, and providing
data reporting on which specific wardrobe items, item styles or
item colors are either similar or identical.
6. The method of claim 1, further providing at least some of said
item images by using said client computerized device to upload
optical codes or catalog information pertaining to said user
wardrobe items, and using said optical codes or catalog information
to retrieve said item images from at least one database of item
images; or wherein said client computerized device further
comprises at least one software controlled camera, further using
said client computerized device to photograph at least one of said
items, thereby producing item images, and uploading said item
images to said server.
7. The method of claim 1, wherein said client computerized device
further comprises at least one camera and a display screen; further
generating a template of at least one item on said display screen,
and instructing said user to use said template to as a photographic
guide while photographing said item, thereby producing template
delineated images; and using only portions of said template
delineated images that fall inside of said template for said item
images.
8. The method of claim 1, wherein said user clothing information
further comprises potential user clothing purchases; further
receiving said potential user clothing purchases, and receiving
social network fashion preferences comprising preferences of at
least one different social network linked contact pertaining to
said potential user clothing purchase; and using a potential
purchase scoring algorithm that operates according to a function of
said social network fashion preferences to give feedback on a merit
of said potential user clothing purchase to said user; wherein said
potential user clothing purchases may be obtained from at least one
of said user, at least one different social network based contact,
clothing advertiser, clothing retailer, or clothing
manufacturer.
9. The method of claim 1, further using wardrobe items from at
least different social network linked contacts, said user's
wardrobe items, and said user fashion preferences, to determine
popular wardrobe items not owned by said user and recommend these
items to either said user or to at least one different social
network linked contact.
10. The method of claim 1, further receiving information pertaining
to a present or future location of said user, and anticipated
weather conditions proximate to said present or future location of
said user; further analyzing said user wardrobe items or outfits
according to at least one of item type, item color, or item style,
and further ranking said wardrobe items or outfits by weather
suitability; wherein said scoring algorithm also gives weighting to
the weather suitability of said wardrobe items or outfits.
11. The method of claim 1, further receiving information pertaining
to at least one event type; further analyzing said user wardrobe
items or outfits according said at least one event type, and
further ranking said wardrobe items or outfits by at least one
event suitability; wherein said scoring algorithm also gives
weighting to at least one event suitability of said wardrobe items
or outfits.
12. The method of claim 11, further receiving calendar information
pertaining to the calendar scheduling of said at least one event
types; further suggesting at least one user wardrobe item or at
least one outfit for said user to wear for each of said at least
one event types according to said calendar information.
13. The method of claim 1, further tracking wardrobe items in at
least said user social networks as a function of time, and
reporting statistics on trends in said wardrobe items as a function
of time.
14. The method of claim 13, wherein said user has a user geographic
location, and wherein said server further comprises trend
prediction software, said trend prediction software configured to
use said user fashion preferences, social network fashion
preferences, said trends in said wardrobe items as a function of
time, and sales data pertaining to clothing sales in said user's
geographic location to predict latest fashion trends with respect
to wardrobe item types, item colors, and item styles.
15. The method of claim 1, either allowing members of said user
social networks to vote on which member has a best fashion sense,
and reporting these results; or scoring mutual social network
fashion preferences among members of said user social network, and
automatically ranking a plurality of members according to their
overall cumulative fashion sense score.
16. The method of claim 15, further providing an advice exchange
forum for all users, wherein said all users may direct questions to
either all users, specific users, or users with a specified overall
cumulative fashion preference score; and wherein any answers may be
directed to either all users, specific users, or the user who
submitted the question; and wherein said user who submitted the
question may rank the quality of said answer according to a quality
score, thus awarding answer points to the answering user who
submitted said answers; and wherein said server may award prises to
answering users with high cumulative answer points.
17. The method of claim 15, further providing fashion games
methods, said fashion games methods being at least one of: a) using
a plurality of client computerized devices equipped with
accelerometers or buttons to determine which said users are
presently activating their client computerized devices by shaking
or pressing at least one of said buttons; and for all users with
presently activated client computerized devices, sharing
information pertaining to said user wardrobe items; b) receiving,
from said users, guesses at what other members of said user's
social networks are wearing, and reporting on a score of an
accuracy of said guesses; c) providing a virtual fashion carnival
by inviting a plurality of other members to exhibit their clothing
selections as either user wardrobe items, user outfits, or
customized virtual mannequins, providing access to said virtual
fashion carnival to viewers comprising either members of said
user's social network or members of said social network and
individuals outside of said user's social network, and receiving
fashion rating votes from at least some of said viewers; wherein
said fashion rating votes are added to said member's said overall
cumulative fashion sense score.
18. The method of claim 1, further receiving and storing in said
database, information pertaining to daily user choice of wardrobe
items and outfits; using said information pertaining to daily user
choice of wardrobe items and outfits to construct a history of said
user's daily user choices; and using said history of said user's
daily user choices in said scoring algorithm.
19. The method of claim 1, wherein said client computerized device
comprises at least one processor, memory, and a display, and
wherein said client computerized device uses either fashion client
software or a web browser to send and receive information from both
said at least one server and said user.
20. The method of claim 19, wherein said client computerized device
is a smartphone or tablet device further comprising a video camera,
wherein said display is a touch sensitive display, and wherein said
fashion client software is a downloadable app.
Description
CROSS REFERENCE TO RELATED APPLICATIONS
[0001] This application is a continuation of U.S. patent
application Ser. No. 14/476,984, "SYSTEM AND METHOD FOR PROVIDING
AUTOMATED CLOTHING FASHION RECOMMENDATIONS, filed Sep. 4, 2014, the
entire contents of which are incorporated herein by reference.
BACKGROUND OF THE INVENTION
[0002] 1. Field of the Invention
[0003] This invention is in the field of automated clothing and
fashion recommendation software
[0004] 2. Description of the Related Art
[0005] Fashion--choice of what types of clothing to wear, and what
not to wear, is extremely important in our society and indeed in
most societies throughout time. An individual's choice of clothing
allows others to almost instantly determine an individual's social
status, outlook on life, and identification with certain socially
important groups. Clothing choices are particularly important to
young adults, who often spend much time and energy trying to both
establish their individual uniqueness, while at the same time
trying to also show what social group and values the young adult
identifies with. The consequences of making wrong clothing choices
for a given event and situation can often be very distressing.
[0006] As a result, the fashion industry is a very large industry,
employing over 4 million people in the US alone, and in the US
about 250 billion dollars a year are spent on fashion
purchases.
[0007] Some individuals may be blessed with a high innate sense of
fashion and style. However many other individuals may have less
innate fashion sense or confidence, but still may wish to dress
well for various occasions. As a result, numerous fashion magazines
and websites exist, and fashion is often a frequent topic of
discussion among friends. Further, with the proliferation of
smartphone technology, various types of fashion "apps" are now
available for download.
[0008] For example, a search for "fashion" apps on the Apple App
store shows that various applications are available, including
"Social Girl" by Crowdstar Inc., "Fashion Design World", by
nanobitssoftware.com, "Kim Kardashian Hollywood", by Glu Games,
Inc., "Forever 21", by Forever 21, Inc., and others. However such
games are often more fantasy oriented, and less oriented towards
helping to solve practical problems of an individual should
actually wear, with their existing wardrobe, in a real world
environment.
BRIEF SUMMARY OF THE INVENTION
[0009] The invention is based, in part, on the insight that what is
needed is a practical social network based clothing recommendation
system. This system should be based on the user's actual clothing
wardrobe, group of real life friends (real-life social network
group), and real-life environment, and try to optimize among these
various constraints to provide high quality fashion advice, and
specific recommendations as to what to wear on any given day or
occasion.
[0010] In some embodiments, the invention may be a client-server
social-network type system and method for providing automated
clothing suggestions. Here at least one Internet server, configured
with the invention's fashion social network software, will interact
with various fashion interested users and create various social
network groups. Within each social network group, the various users
will typically connected because they are friends (or friends of
friends), and also because they share certain fashion values in
common. That is, unlike prior social networks, group members will
typically share common fashion ideals. Within each group, members
will typically be encouraged to pick friends who share similar
tastes in fashion. The users will typically interact with the
server using their camera equipped smartphones, or other
computerized devices.
[0011] The server's fashion social network software will further be
configured to store detailed records of each user's wardrobe, and
for each wardrobe item will often store a description of the
wardrobe item type, style, color, designer, material, occasion
suitability, weather suitability user preference, history of being
worn by the user, and the like. Users will additionally be able to
combine various wardrobe items into outfits composed of multiple
wardrobe items, and also visualize these individual wardrobe items
and outfits by use of virtual mannequins and the like. Users will
also be able to share their wardrobe items, outfits, and virtual
mannequins with other users.
[0012] The server's fashion social network software will
additionally be configured to allow other members of a user's
social network group to evaluate their own wardrobe items and
outfits, as well as wardrobe items and outfits belonging to their
friends, and friends of friends, and to draw statistical inferences
as to what types of clothing are favored and disfavored by that
group. These statistical inferences can then be used by the
invention's scoring algorithm to produce recommendations as to what
items and outfits a user should wear. Here the scoring algorithm
may, for example, weigh various factors as to the user's own
preferences, how recently the user's various wardrobe items were
last worn, and the user's social network group fashion
inclinations. The scoring algorithm will then recommend clothing
choices that are compatible with these various factors. In more
advanced embodiments, the system's scoring algorithm may
additionally factor in other variables, such as the user's location
and the weather at this location, weather, type occasion or event
(which may be based on calendar schedules), and the fashion
preferences of social network users outside of the user's group as
well.
[0013] In addition to making "what to wear" clothing
recommendations, the system can additionally assist in making
shopping or gift recommendations, as well as to provide various
fashion related games, discussion boards, and interface various
clothing suppliers to provide more advanced fashion trend and
analysis services to the clothing suppliers. Indeed, the system
could be a valuable research tool for the fashion industry, and at
least some of the system operating expenses could be paid by
providing such research tools to the fashion industry.
BRIEF DESCRIPTION OF THE DRAWINGS
[0014] FIG. 1 shows an overview of the client-server structure of
the invention's automated clothing suggestion system. Typically at
least one or more Internet servers are configured to implement a
social network for a plurality of users operating a variety of
different client computerized devices (often Smartphones, tablet
computers and the like). The typical user will often be a young
adult, interested in clothing and fashion styles. The typical user
will often have a desire to confirm that their choice of wardrobe
items and clothing outfits conforms to styles for their particular
social group of friends, friends of friends, and the like. Here
some of these various social groups of friends and friends of
friends are shown.
[0015] FIG. 2 shows a graphic depiction of some of the clothing
data and data structures that can be stored in the Internet
server's database. Here, for each user, the server's database will
store data and images for a plurality of wardrobe items, often
indexed by item type (e.g. dresses, skirts, pants, belts), item
style (e.g. long skirts, short skirts, long pants, short pants),
item color, item image, item history information (e.g. when and
where purchased, item bar codes or identification numbers, and the
like) and other data to be discussed. The server database can also
contain information to construct at least one virtual 2-dimension
or 3-dimensional mannequin, such as at least one image of the
virtual mannequin's face, user body dimensions, and the like. The
user(s) can use their computerized devices to access the server,
send and receive data, and scroll through the various wardrobe
items and receive recommendations, and also view outfits (of
multiple wardrobe items) on their virtual mannequin(s).
[0016] FIG. 3 shows how a user can add additional wardrobe items to
their server database, and/or update one or more images (such as
images of the face) used for their particular virtual mannequin.
Here the user is using a camera on their client computerized device
(such as a Smartphone) to either take pictures of a new wardrobe
item (long pants) or alternatively scan a bar code or other
wardrobe item identification information. This information is then
uploaded to the server database. In FIG. 3, the user is also taking
an image of their own face and uploading this image to the server
database, so that their virtual mannequin will have a face similar
to the user's own face. The user may also upload customization data
for other aspects of the virtual mannequin (e.g. height, weight,
various body dimensions) as well.
[0017] FIG. 4 shows how various members of a particular social
group of friends and friends of friends can enter in their
particular favored wardrobe items or outfits that they have
previously worn, are presently wearing or are planning to wear, and
send this to the server. The server can use this information, along
with information pertaining to a particular user's wardrobe and
particular user fashion preferences, to suggest items of clothing
(here long light gray pants) that generally fit in with the
preferences of other members of the user's social group, yet may
not be identical to the other outfits.
[0018] FIG. 5 shows how the system can also help detect wardrobe
"collisions" where two members of the system may be planning to
attend the same event at the same time, yet do not want to wear
identical outfits. Here the system has detected that the user's
proposed outfit is identical to an outfit that another member of
the same social group (or alternatively any social group in the
social network) has previously planned to wear, and is thus warning
the user about the likely outfit "collision". By default the
system's collision detection algorithms may often operate on a
"first to claim an outfit" priority basis, but other priority
schemes may also be implemented.
[0019] FIG. 6 shows more detail pertaining to the various methods
used by the system to prevent clothing "collisions", where more
than one user will wear the same outfit or major wardrobe item
(such as the same dress) to the same event.
[0020] FIG. 7 shows an example of one type of fashion game. Here
the server is configured to detect whenever any server user is
shaking their computerized device, and then show an image of the
user's virtual mannequin wearing the user's present outfit to those
other users who are shaking their devices at the same time. In this
example, this sharing operates regardless of any social group
compartmentalization, and may help one social group find another
social group with similar fashion sense. Other game configurations
(e.g. only share within a social group, or only within social
groups of the same "social group type") may also be
implemented.
[0021] FIG. 8 shows an example of a different type of fashion
game--here a "virtual fashion carnival" in which members of a
particular social group have arranged to combine their favorite
outfits, show outfits (one from each participating user) on screen
at the same time (here using the user's various virtual
mannequins), and solicit ratings of these outfits. Here, in an
attempt to get more objective ratings, the members of this
particular social group have elected to solicit ratings from other
social groups on the social network.
[0022] FIG. 9 shows that in some embodiments, the system can use
relatively complex methods to automatically (or adaptively) adjust
the scoring algorithm, as well as to compute user and social
network fashion preferences to better compensate for fashion trends
in a given region. These complex methods can also be useful for
generating fashion data and analysis for various reports. In this
embodiment, the regional clothing style trend prediction methods
can integrate data from multiple sources, such as user personal
preferences, regional store sales data, forum discussions, social
network friend ranking data, news feed data, geographic data, and
the history of preferences and daily outfit choices from different
users in the same region. The system can then take this data, and
automatically compute statistical reports pertaining to the latest
fashion trends, such as what types and styles of clothing, in what
colors, are becoming more popular.
[0023] FIG. 10 further shows that in some embodiments the scoring
algorithm can also integrate raw data from multiple sources,
including sources outside of the social network. The system can
also do various types of automated data mining, and in turn use
this information to guide the scoring algorithm.
DETAILED DESCRIPTION OF THE INVENTION
[0024] FIG. 1 shows an overview of the client-server structure of
the automated clothing suggestion system. Typically at least one or
more Internet servers (100) are configured to implement a social
network with a plurality of users running a plurality of user
client computerized devices (often smartphones, tablet computers
and the like) (112 . . . 118, 122 . . . 126, and 132 . . .
135).
[0025] The typical user will be interested in clothing and fashion
styles, and will often have a desire to confirm that their choice
of wardrobe items and clothing outfits conforms to styles for their
particular social group of friends, friends of friends, and the
like. Here some of these various social groups of friends and
friends of friends are shown. For example, group (110) may
generally consist of friends who mostly are 8.sup.th grade students
at school "a", group (120) may generally consist of friends who
mostly are 9.sup.th grade students at nearby school "b", and group
(130) may consist of friends who mostly are 10.sup.th grade
students at different school "c", and so on.
[0026] Put alternatively, in some embodiments the invention may be
client-server social-network system and method for providing
automated clothing suggestions. This system and method will
generally comprise using at least one server (100). This server
will typically comprise at least one processor (often of the
popular ARM, x86, MIPS, PowerPC, or other processor family),
memory, operating system and often web server software, and fashion
social network software. This invention will typically be
implemented using this fashion social network software. The fashion
social network software itself may often be implemented using
various LAMP stack software, such as Linux, Apache web servers,
MySQL (or MariaDB, MongoDB, etc.) type database software, and
programming languages such as PHP (or Perl, Python, etc.) and the
like, but other types of software from other suppliers may also be
used.
[0027] The server (100) will also usually have a network connection
to the internet (104), and the fashion social network software,
when running under the control of at least one server processor,
will automatically provide the framework for a fashion social
network. This fashion social network will comprise a plurality of
users (here exemplified by their client computerized devices (112 .
. . 118, 122 . . . 126, and 132 . . . 134), as well as a database
(102). This database (102) will typically be configured to store,
on a per user basis, user clothing information, user fashion
preferences, user social network linkages, and social network
fashion preferences. Users will typically form groups where group
members typically are related by a "friends" or "friends of
friends" basis, and due to the fashion orientation of the system,
groups will be encouraged to select primarily members with similar
fashion taste. Here, for example, a group may optionally require
new potential members to answer a fashion quiz (either provided by
the system, or provided by the members themselves) to help insure
similar fashion interests.
[0028] Groups may thus have fashion themes, and one user may be a
member of more than one group on this fashion theme basis. The
social network as a whole will thus comprise multiple groups, such
as (110), (120), and (130). Certain "super groups" where users may
not all have a "friends" or "friends of friends" relationship, but
may have a common fashion sense or interest in common fashion
themes, may also be established.
[0029] Some of the user clothing information is shown in more
detail in FIG. 2. This user clothing information will comprise
various user wardrobe items (210, 220, 230, 240, 250) that the user
presently owns or has access to, such as (for girls) blouses,
pants, belts, skirts, dresses, and the like. Each user wardrobe
item will further comprise item type (e.g. blouses, pants, belts,
skirts, dresses, and the like), item colors, item style(s) (e.g.
V-neck blouses (212), non-V-neck blouses (214), long pants (222),
short pants (224), thick belts (232), long skirts (242), short
skirts (244) and the like. The database may also contain item
images (e.g. computer format images or drawings of each item), and
other item information such as item history information, item
weather suitability, item occasion suitability, user preference for
the item, social group preference for the item, and the like.
[0030] According to the system and method, at least one server
(100) will be further configured to receive information pertaining
to any given user's clothing information (see FIG. 3 for more
detail), fashion preferences, and user friend information (e.g. the
friends of any given user) from the user's client computerized
device. The client computerized devices, exemplified by devices
(112 . . . 118, 122 . . . 126, and 132 . . . 134) in FIG. 1 and
elsewhere will generally themselves comprise at least one
processor, memory, and a display. The client computerized devices
will use either specific fashion client software, or a web browser,
to send and receive information from server (100). Here desktop
computers, laptop computers, tablet computers and smartphones and
the like may be used as client computerized devices.
[0031] In a preferred embodiment, the client computerized devices
may be smartphones or tablet devices that further comprising at
least one video camera, a touch sensitive display, and a wireless
network connection (e.g. Wi-Fi connection, cellular telephone
connection, and the like) that ultimately extends to the Internet.
These types of devices may run the fashion client software is a
downloadable app. Examples of this later type of devices may be
Apple iOS iPhones, Android smartphones, Windows smartphones, and
the like.
[0032] The server (100) will receive the information from the
client computerized device, and store this information (e.g. the
user's clothing information, and preferences and friend
information) in database (102). The system server (100) and
database (102) can then use this user friend information to
establish or maintain various social network linkages between the
various users, thereby establishing user social networks.
[0033] For example, returning to FIG. 1, let us focus on the user
of client computerized device (112)). User (112) enters in the
information that she has friends (users of devices 114 and 116),
and the system can then determine that these three friends must
therefore form social group (110). Further, friend (116) in turn is
friends with the user of device (118), and thus the user of device
(118) also joins social group (110) and has a friends-of-friends
relationship with the owner of device (112). Assume at present that
all members of the group have generally similar fashion ideals.
Other social groups such as (120) and (130) may also form in this
manner.
[0034] As will be described in more detail, the system and method
operates by further receiving, often on a per-user basis, social
network fashion preferences from the user social network. Here
depending upon system settings, these fashion preferences may be
restricted to group (110), or may be set to be broader and to pool
data from various other social groups (ideally from groups with
similar overall fashion values), such as (120) and (130), thereby
allowing for larger fashion "super groups" to be formed when this
option is desired.
[0035] For example, these social network fashion preferences can
comprise the preferences of at least one different social network
linked contact (e.g. for user 112, this would be the preferences of
at least users 114, 116 or 118, and preferably at least several of
these) pertaining to at least one user wardrobe item (e.g. at least
one of user 112's blouses, pants, belts, skirts, or dresses in FIG.
2), or at least one outfit (e.g. blouse, pants, belt, or dress, or
blouse skirt, belt combination, etc.) comprising a variety of
different user wardrobe items.
[0036] The server (100), at least one server processor, and the
server's fashion social network software can use this user clothing
information, user fashion preferences (e.g. user 112 can transmit
information to the server regarding favorite wardrobe items,
preferred styles, and the like), and the social network fashion
preferences (e.g. preferences of other social network members) and
a scoring algorithm to suggest at least one user wardrobe item, or
at least one outfit, for the user to wear.
[0037] See for example, FIG. 2, where the user is using her device
(112a) to perhaps scroll through the user's wardrobe of pants items
(220). Here the system is recommending the best pair of pants to
wear (here long pants 222b) by the arrow (270). (The logic behind
this decision is shown in more detail in FIG. 4.) Alternatively the
system may recommend an entire wardrobe (combination of pants 222b,
blouse 214a, and belt 232c) as shown in FIG. 2 (112b).
[0038] As will be described in more detail, this scoring algorithm
can give higher or lower weights to various user (112) wardrobe
items or outfits according various criteria. These can include
criteria such as (a function of) the social network fashion
preferences (e.g. preferences of at least users 114, 116, 118 in
user 112's social group 110), user 112's own fashion preferences,
and the item history information (e.g. did user 112 just wear that
item or outfit yesterday? if so then give it a lower weight). In
other words, the item history information can be the user's history
of recently wearing that item or outfit. Usually more recently worn
items will be given lower weight to avoid over repetition, but the
user may choose to adjust this weighting according to their
individual desires (here the system may optionally warn if a choice
is overly non-standard).
[0039] In operation, server (100) and database (102) will generally
further receive information pertaining to an actual wardrobe item
or outfit choice of the user. That is, if user (112) accepts the
FIG. 2 pants suggestion (270) of pants (222b), user (112),
generally prompted by software running on board user (112's)
device, will inform the server that these where the pants that the
user actually used. Alternatively the user may inform the server
that the user ended up wearing some other selection. The server
will then store this "actual choice" information in its database
(102), and use the user's actual choice as at least part of the
user's user fashion preferences in subsequent recommendations.
[0040] FIG. 2 also shows a graphic depiction of some of the
clothing data and data structures (200) held in the Internet
server's database (102). Here, for each user, the server's database
(102) will store data and images for a plurality of wardrobe items,
often indexed by wardrobe item clothing type (e.g. dresses (250),
skirts (240), pants (220), belts (220) and blouses (220)), item
style (e.g. long skirts (242), short skirts (244), long pants
(222), short pants (224)), item color, item image, and item history
information (e.g. when and where purchased or worn, item bar codes
or identification numbers, and the like). The server database (102)
can also, for each user, contain information (260) to construct a
2-dimensional or 3-dimensional virtual mannequin (264), such as at
least one image of the virtual mannequin's face (262). The user(s)
can use their computerized devices (e.g. 112) to access the server
(100), send and receive data, and scroll through the various
wardrobe items (112a) and receive recommendations (270), and also
view outfits (of multiple wardrobe items) on the virtual mannequin
(112b). In some more advanced embodiments, the virtual mannequin
may be animated by the system and may pose, walk down a virtual
runway, turn around, etc., and the wardrobe item images will be
used as surfaces, skins or textures, on the moving virtual
mannequin (often a 3-dimensional moving mannequin), and can be
manipulated using standard 3D graphics techniques to realistically
show what the clothing would look like under conditions of 3D
movement.
[0041] Thus for example as shown in FIG. 2 (112a), the user can
thus further sort their user various wardrobe items or outfits
according to various categories, such as item types (e.g. general
type of clothes), item colors (color of clothes), item styles (e.g.
long, short, formal, informal, etc.), item history information
(name of manufacturer or designer, when last worn, etc.), user
fashion preferences (e.g. prefer formal, prefer casual, prefer
light colors, prefer dark colors, etc.), and social network fashion
preferences (e.g. preferences of at least the members of social
group 110 in the case of user 112), and the system can then produce
sorted user wardrobe items or sorted user outfits. Images of these
various items (e.g. the images of the various long pants in (112b))
can be used to display these sorted user wardrobe items (or sorted
user outfits, see 112b) on the relevant client computerized device
(e.g. 112).
[0042] In particular, in FIG. 2, note that the system can use the
client computerized device to display the wardrobe item or outfit
on a virtual mannequin. As will be shown in more detail in FIG. 3,
the system can also allow the user to upload images (e.g. faces,
such as the user's face 304) to the server (100) and use these
images (262a) to customize the appearance of the virtual mannequin
(262a) to resemble the user. This thus allows the user to make a
customized virtual mannequin, and display the user wardrobe items
or outfits (often sorted according to the system's scoring
algorithm) on the user's client computerized device.
[0043] As previously discussed, FIG. 3 shows how a user can add
additional wardrobe items (300) to their server database (102,
200), and/or update one or more images (such as images of the face
262a) used for their particular virtual mannequin (264a). Here the
user is using a camera on their client computerized device (112c)
(such as a smartphone) to either take pictures of a new wardrobe
item (long pants 300) or alternatively scan a bar code (302) or
other wardrobe item identification information for this new
wardrobe article of clothing. This information is then transmitted
by device (112c) over the Internet (104) to the server (100) and
uploaded to the server database (102). FIG. 3 also shows an example
where the user is also taking an image of their own face (304) and
uploading this image to the server database (102, 200) so that the
virtual mannequin (264a) will have a face (262a) similar to the
user's own face. The user may also upload customization data for
other aspects of the virtual mannequin (e.g. height, weight,
various body dimensions) as well.
[0044] In some cases, the user may wish to show their virtual
mannequin or their various wardrobe items against a customized
moving or static background image or video. For example, a user
intending to go to a formal event may wish to use a formal event
type background image, while a user intending to go to a beach
party may wish to use a beach type background image. To do this,
the system can also be configured to allow the user to further
select or upload background images, video, (or a link to these
images or dynamic background) to the server database (102) for the
virtual mannequin to pose against. Alternatively the system may
provide various standard backgrounds to use. Once the background is
chosen, the system can then display the items of interest (e.g.
sorted user wardrobe items or sorted user outfits) on their
customized virtual mannequin against these various user chosen
background images or dynamic background.
[0045] Again, as previously discussed in FIG. 3, in order to let
the system (e.g. server 100, database 102) know what specific
wardrobe items the user has (or wants) and what these items look
like, the user has various options. In one option, the user can use
their client computerized device (e.g. 112) to photograph or
otherwise scan optical codes (302) (e.g. 1D or 2D bar codes),
catalog numbers, or other information (e.g. name of item,
manufacturer), RFID tag data, etc. pertaining to the wardrobe item
of interest, and upload this information. The server can then use
this uploaded information (e.g. optical bar codes, catalog numbers,
and the like to then retrieve images and other information
regarding the item from another database, such as a manufacturer or
catalog database. For such purposes (e.g. scanning optical bar
codes), it is useful if the client computerized device (112) has at
least one software controlled camera. If so, then the user can
simply use the client computerized device's camera to photograph
the appropriate bar codes, wardrobe item itself, or other
information, and upload the images to the server (100).
[0046] In the event that the user wishes to simply take a
photograph of the wardrobe item for uploading, various methods may
be used to help ensure that good quality images result. For
example, in a common situation where client computerized device has
at least one camera and a display screen (e.g. a smartphone or
tablet computer), then the device software may be configured to
inquire as to what type of wardrobe item is being photographed. The
device can then use this information to, for example, generate a
template of that particular item on the display screen, as well as
to instruct the user to use this template to as a photographic
guide. This example is shown in FIG. 3 (112c). Here the system has
generated a photographic guide or template of the long pants (300)
that the user wishes to photograph and add to her wardrobe database
(102, 200). The user can use this template while photographing the
items, generally moving the camera to fill the screen image of the
template with the image of the clothing item. The system can then
only use the portions of the image that fall within the template to
generate the image of the wardrobe item. Here the uploaded
photograph of the long pants will be added to the database as
(222d).
[0047] FIG. 4 shows how various members of a particular social
group of friends and friends of friends (110) can enter in their
particular favored wardrobe items or outfits that they may be
wearing or planning to wear (here 114a, 116a, and 118a), and send
this to the server (100). The server can use this information,
along with information pertaining to a particular user's wardrobe
(e.g. user 112's wardrobe, previously shown in FIGS. 2 and 3) and
particular user fashion preferences, to suggest items of clothing
(here long light gray pants 270) that generally fit in with the
preferences of other members of the user's social group (e.g. all
like pants, with a general preference for long pants), yet may not
be identical to the other outfits (e.g. the color of the
recommended pants (270) is light gray, midway between the long dark
gray pants worn by user (114a), and the long white pants worn by
user 118a).
[0048] As previously discussed, in a preferred embodiment, the
system will further be configured to receive information pertaining
to the various users' actual daily choice of wardrobe items and
outfits (e.g. what the users actually end up wearing). Here for
example, the system can use the various client computerized devices
to prompt for this information, or otherwise encourage the users to
enter in their final choices. These final choices can then be
stored in database (102). The system can then use this information
(pertaining to daily user choice of wardrobe items and outfits) to
construct a history of the user's daily choices in what they
actually wear. This information can then be used for various
purposes, including using this history in the previously discussed
scoring algorithm. One immediate use for this type of data is to
help prevent two different users from showing up at the same event
wearing wardrobe items or outfits that are perceived as being "too
similar", in other words, a clothing or wardrobe "collision".
[0049] FIG. 5 shows how the system can help detect and prevent
wardrobe "collisions", where two members of the system may be
planning to attend the same event at the same time, yet do not want
to wear identical outfits. Here the system has detected that user
(112c)'s proposed outfit (is identical to an outfit earlier chosen
by another member (114a) of the same social group (110) (or
alternatively any social group in the social network). The system
here is warning the user about the likely outfit "collision", by
drawing the "X" mark (500) over user (112c)'s proposed outfit (here
any type of warning, including audio messages, text messages
alternate symbols, etc. may be used). By default the system will
often operate on a "first to claim an outfit" basis, but other
priority schemes may also be implemented. In theory a "queen bee"
type social group member with high status might, upon consent of
the other members, be given an overriding priority in her choice of
outfits.
[0050] This collision detection method can work, for example, by
further receiving (at the server 100) the present or reported
future locations of the user (e.g. user 112c) and at least some
individuals in at least the user's social group (110) or, depending
upon the setting, other individuals in other groups in the social
network such as, for example, groups (120) and (130). Here for
example, the user may inform the system that they are going to
their school's Saturday night dance (assume that the school
location is on file with the social network).
[0051] In this embodiment, the collision detection scheme works by
determining if the present or reported future locations of the user
and at least some individuals in the user's social group or
network, are also planning to be within the same given location
criteria (e.g. also going to their school's Saturday night dance).
The server (100) can use its at least one processor to determine if
either user (112)'s present or reported future locations, as well
as at least some individuals in the user's group or social network,
are or plan to be at the same location at the same time (e.g.
within a given location error criteria, such as matching the
location within +/-100 feet of the location, or other error range
that can be set by the system or alternatively be adjusted by the
users).
[0052] Assuming that the system determines that two or more
individuals from the same social network plan to be at the same
location and place, the server can then perform various additional
checks: For example, the system may receive information pertaining
to what wardrobe items or outfits that the various individuals
(e.g. user (112) and other) are wearing or plan to wear, and
determine if the outfits match, or major wardrobe items (e.g.
dresses) match, or are the same color or otherwise meet certain
preset collision parameters. These preset collision parameters may
be often determined by system defaults, (e.g. warn on if both top
and bottom wardrobe items are the same, warn if dresses are the
same, etc.), or alternatively these preset collision parameters may
be modified by the users (e.g. user set to warn if the same belts
are being worn). The sever (100) can then provide a collision
warning if at least some of the item styles or item colors are
either similar or identical, as per the preset collision
parameters.
[0053] Alternatively or additionally, the system can also give more
collision information than just the large "X" (400) shown in FIG.
5. For example, the server can also report more information, such
as providing specific wardrobe item data reporting on which and how
many of the wardrobe items, item styles, or item colors are in
collision between the various users.
[0054] FIG. 6 shows more detail pertaining to the various methods
used by the system to prevent clothing "collisions", where more
than one user will wear the same outfit or major wardrobe item
(such as the same dress) to the same event.
[0055] Fashion Ranking, Advice, and Games:
[0056] In some embodiments, the system can also be configured to
allow members of any given user's social group (e.g. members of
group 110), as well as other members of other social network groups
(which can be members of different social groups such as 120 and
130) to vote on the fashion sense of the various users/members.
Here for example, users can vote on which other users/members have
the best fashion sense. Other types voting for other types of
fashion ranking schemes can also be used, such as numerically
ranking the various user/members fashion sense, voting on the
user/member with the worst fashion sense, and so on. The system can
then report these voting results, and automatically rank the
various members in a particular voting group or poll according to
their overall cumulative fashion sense score.
[0057] In some embodiments, the system may also run various types
of advice exchange forums (message boards, online discussion
systems), in which users can write in with fashion questions and
observations, and these can in turn be commented on by other users.
However because not all comments will be from individuals with
equal fashion sense, it may be useful to also configure thee system
to provide a mechanism by which comments or posts from individuals
with a greater level of peer acknowledged fashion sense can be
distinguished from comments and posts from individuals with lower
levels of peer acknowledged fashion sense.
[0058] Here, the previously discussed fashion sense voting or
ranking system can be useful for these purposes. In this
embodiment, the system can provide an advice exchange forum
(message forum, online discussion form) for some or all users
(either unrestricted, or on a user group or super group basis).
Here the various users may direct questions to, for example, either
all users, specific users, or users with a specified overall
cumulative fashion preference score. Answers to these questions in
turn may be directed to either all users, specific users, or the
user who submitted the question. The relative cumulative fashion
preference score of either the user submitting the question and/or
the user answering the question may also be listed so that the
relative merits of the question and answer may also be assessed by
other individuals.
[0059] In some embodiments, the online discussion system may be
further configured to also allow users to rank the quality of
answers. For examples, users may vote on the quality of various
answers according to various types of quality scores (e.g. like,
dislike, 1-5 rating, A-F ratings, and so on). Persons who submit
answers that are judged by other users as being of high quality may
be given more answer points. The system may even be configured to
reward message board participation by awarding prizes to
individuals who with a high cumulative number of answer points.
[0060] In some embodiments, the system (e.g. server 100) may also
be configured to provide various types of fashion games. Here
examples of three games: a "shake and compare clothing" game, a
"virtual fashion carnival" game, and a "guess what I am wearing"
game are presented.
[0061] The "shake and compare" game takes advantage of the fact
that many users will be using smartphones to access the server, and
that most modern smartphones are equipped with accelerometers that
can detect shaking. Note, however this game can also be implemented
by other methods, such as by pushing a button on the client
computerized device, as needed or desired by the user(s). Here,
various social network users, either within a given social group
(110), or between different social groups (e.g. 110, 120, 130) may
be prompted to, at a given moment, shake their client computerized
devices (smartphones) or push a button on their various client
computerized devices. When this happens, at least for those client
computerized devices that are presently activated by shaking or
button pressing, then the server can then share information
pertaining to the wardrobe items that these particular users are
presently wearing. For example, the server (100) can transmit
images of the user's virtual mannequins, dressed in the wardrobe
items that the users are presently wearing, to the screens of the
various activated client computerized devices, and the users can in
turn scroll through or otherwise inspect what other users with
activated client computerized devices are wearing at any moment in
time.
[0062] FIG. 7 shows an example of this "shake and compare" game. As
discussed above, server (100) may be configured to detect whenever
any social network user is shaking their computerized device (here
the users of devices 116a, 124a, and 132a are presently shaking
their devices), and then show the user's outfits (here on the
various user's virtual mannequins) to the other users who are
shaking their devices at the same time. In some game embodiments or
settings, this can be done regardless of any social group
compartmentalization. Here for example, users (116b) (124a), and
(134a), who are normally in different social groups (110), (120),
and (130), can nonetheless communicate to at least this limited
extent. In other game embodiments or settings, this "shake and
compare" game can be configured so that only social network users
within a given social network group (e.g. within group 110) or
super group can exchange wardrobe information when they
simultaneously shake or otherwise activate their devices.
[0063] In an alternative "virtual fashion carnival" game, the
server (100) (and client computerized devices) can be configured so
that the users (often within a social group such as 110) can invite
various other social network members (either within the same social
group 110 that is providing the virtual fashion carnival, or with
members of different social groups or super groups such as 120 and
130 on the same server social network) to exhibit their clothing
selections. These clothing selections can be exhibited as either
specific user wardrobe items, or user outfits (composed of multiple
wardrobe items) often worn by customized virtual mannequins.
[0064] The users (fashion carnival presenters) putting on the
virtual fashion carnival can then provide access (through server
100) to the virtual fashion carnival to various viewers (e.g.
members of the fashion carnival presenter's social network group
such as 110, or members of both the presenter's social network
group 110 as well as various individuals outside of the fashion
carnival presenters social network group, such as members of the
fashion super group). Additionally, the server can also be
configured to receive fashion rating votes from at least some of
the various viewers, ranking the outfits or wardrobe items of the
various fashion carnival presenters according to their fashion
sense. The system can also be configured to encourage users to
conduct such virtual fashion carnivals by, for example, adding
these fashion rating votes to the fashion carnival presenter
member's overall cumulative fashion sense score.
[0065] This virtual fashion carnival is shown in more detail in
FIG. 8. Here the members of a particular social group (e.g. group
110 users who control devices 112d, 114b, 116c and 118b) have
arranged to show a virtual fashion runway presenting the fashion
carnival presenter's favorite outfits, disposed on virtual
mannequins, on a show that can be seen by various other
individuals. The presenters have also decided to request that the
other viewers rate the presenter's outfits. Here the user who
controls device 122a, from different social group 120, likes the
outfit in front and is voting favorably (up arrow). By contrast the
user who controls device 132a, from different social group 130,
does not like the outfit in front and is voting unfavorably (down
arrow). The server (100) and database memory (102) can keep track
of these voting results and assign them to the overall cumulative
fashion score of that particular fashion carnival presenter, or
otherwise keep track of the voting results.
[0066] The system can also implement various other types of fashion
games as well. For example, another type of fashion game can be
wardrobe item or outfit guessing games. Here those users
participating in the fashion guessing game can, for example, guess
at what other members of the user's social networks (usually social
network group) are wearing. The server (100) and database memory
(102) can keep track of these guesses, as what was actually being
worn, and report results on the accuracy of these guesses.
[0067] Fashion Statistical Reporting Methods:
[0068] In some embodiments, the system may be configured to track
various fashion trends for either the users, and/or for the fashion
industry as well. For example, the system may track various user
wardrobe items as a function of time (e.g. what types, styles, and
colors are presently favored or disfavored, as well as additional
information such as wardrobe item designer or manufacturer name,
fabric or other material type, and the like) across multiple users,
and report statistics on trends as to what is appearing in various
types of user's wardrobes, as well as what various types of users
are actually wearing. The statistical data can also be split out by
various social network groups, so that for example, if group (110)
is starting to favor white short pants as a function of time, while
group (130) is starting to favor black long pants as a function of
time, the different trends of the different groups may also be
identified.
[0069] In some embodiments, statistics and fashion trend reporting
may be further analyzed by user geographic location, and/or
additionally by wardrobe item sales data as well. Here the system
may encourage (or require) that the various social network users
report their main or "home" geographic location (often the school
or university that they attend, home location, or location of work)
to server (100) and database (102). Here the server based fashion
trend prediction software can be configured to track the various
users' fashion preferences, social network fashion preferences, and
report on trends in what wardrobe items are owned, worn, or
purchased as a function of time.
[0070] In some embodiments, the server based fashion trend
prediction software can also use fashion sales data (e.g. data from
manufacturers, retail stores, chains, catalog stores) pertaining to
clothing sales broken down by various categories (such as the
user's geographic location) to analyze and predict latest fashion
trends. These trends can be reported with respect to wardrobe item
types, item colors, item styles, and the like. Because some fashion
businesses may be reluctant to share their private sales data with
other organizations, in some embodiments the system may be
configured to offer a more detailed sales analysis as a premium
confidential service to the clothing industry. In this way, funds
obtained from the clothing industry may in turn help defray the
cost of running the social network system.
[0071] Other more complex scoring algorithm and trend analysis may
also be done. Examples of such more complex methods are shown in
FIG. 9. Here, FIG. 9 shows that in some embodiments, the system can
automatically (or adaptively) adjust the scoring algorithm(s),
compute user fashion preferences, and compute social network
fashion preferences to better compensate for fashion trends in a
given region, as well as to generate data for various reports.
These regional clothing style trend prediction methods can
integrate data from multiple sources, such as user personal
preferences, regional store sales data, forum discussions, social
network friend ranking data, news feed data, geographic data, as
well as the history of preferences and daily outfit choices from
different users in the same region. The system can then compute
statistical data pertaining to the latest fashion trends, such as
what types and styles of clothing, in what colors, are becoming
more popular.
[0072] Additional Methods:
[0073] Shopping and gift assistance: In some embodiments the server
may be configured to further keep track of potential user clothing
purchases (e.g. potential shopping purchases). Here the user(s) can
use their respective computerized devices (or other devices, such
as by relaying from a clothing sales website) to receive and keep
track of the user's potential clothing purchases. Here for example,
a user may research a clothing item for sale before actually
purchasing it. In general the user potential clothing purchases may
be obtained from the shopping user directly, or from least one
different social network based contact (e.g. another user shopping
for a gift), or by relaying browsing information from a clothing
advertiser, clothing retailer, or clothing manufacturer to server
(100).
[0074] The server (100) can receive these user potential clothing
purchases, and also receive other information, such as social
network fashion preferences (e.g. the preferences of at least one
different social network linked contact) pertaining to this user
potential clothing purchase. This can help users make informed
purchases for themselves, and for others as gifts.
[0075] Thus system can, in some embodiments, make use of a
potential purchase scoring algorithm. This algorithm may operate
according to a function of the social network fashion preferences
to give feedback (to the user) on the merit of a user potential
clothing purchase.
[0076] Thus, using FIG. 1 as an example, if the user of device
(112) is shopping for a pink blouse, and the server (100) and
database (102) know (by their record of the wardrobe items of users
114, 116, and 118 in the same social group) that no one owns any
pink blouses, then the system may automatically warn user (112)
that pink as a color is not favored by social group (110).
[0077] Depending upon system settings, the user (112) may choose to
tell the server to also take data from certain other groups (e.g.
groups 120 or 130) (e.g. the fashion super group comprised of other
groups with similar fashion taste) into account as well. The user
can optionally adjust the weight given to input from these other
groups. Thus for example, if pink is not favored by the local
group, but perhaps is favored by other well regarded groups in the
fashion super group (perhaps by certain trendsetter groups), then
the user might decide to disregard the consensus of the local group
in favor of the consensus of other trendsetter groups.
[0078] Similarly if user (118) is shopping for a gift for user
(112), and picks short black pants, the system may analyze user
(112)'s wardrobe and warn user (118) that although user (112) has
various black items, user (112) does not own any short pants, and
this item may be a risky gift.
[0079] To further elaborate, in some embodiments, the system may
further use its knowledge about wardrobe items associated with the
user's various different social network linked contacts (e.g. at
least members of the user's social network group), as well as the
user's own wardrobe items and fashion preferences, to determine if
there are any popular wardrobe items that the user does not
presently own. The system may then recommend these to either the
user (as a potential purchase) or to other members of the user's
social network group (as a potential gift).
[0080] Making Fashion Recommendations Based on Local Weather or
Calendar "Special Event" Considerations:
[0081] In some embodiments, server (110) may be configured to
further receive information pertaining to a present or future
location of a user, as well as the anticipated weather conditions
when the user is at this present or future location. The server can
use this information in a more advanced type of scoring algorithm
to make weather and location appropriate clothing recommendations.
Here for example, the server can further analyze the user wardrobe
items or outfits according to their weather suitability. This
wardrobe item weather suitability may have been previously entered
into the wardrobe item database by user input (e.g. the user tells
the server which wardrobe items are best suited to which weather
conditions). Alternatively the weather suitability of a given
wardrobe item may have been automatically determined by the server
using information from other item data fields such as the item name
(e.g. "raincoat", "light jacket"), or item materials (e.g. plastic,
cotton, leather, suede). As yet another method, when the wardrobe
item is linked to a specific clothing supplier, then the clothing
supplier's own assessment of that wardrobe item's weather
suitability may be automatically imported into server database
(102) and used for these purposes.
[0082] In either event, once the weather suitability of the various
wardrobe items, and the user weather conditions are known, the
system can then adjust its scoring algorithm to also consider (or
weight) the weather suitability of the various wardrobe items or
outfits. This way the system will be less likely to recommend
shorts in subzero weather conditions, or heavy jackets during hot
summer days, and so on. Instead the system will try to make weather
appropriate fashion optimized recommendations.
[0083] Similarly it is often useful to have the system further
optimize its fashion recommendations according to the type of
occasion (e.g. formal, informal, type of occasion, and so on). Here
the server (100) will typically be configured to receive
information pertaining to at least one event type that the user is
attending or planning to attend. The server and associated database
(102) should further be configured so that at least the user,
and/or the clothing supplier (when identified) can further identify
the various wardrobe items as to event type (e.g. occasion such as
formal, informal, sport, dance, beach, etc.). The server (100) can
then further analyze the various user wardrobe items or outfits
according the event type, and as a result thus further rank the
various wardrobe items or outfits by as to how suitable they are
for any given occasion (event). The scoring algorithm can thus take
this occasion (event) suitability ranking into account as well, and
as a result, can make event appropriate fashion optimized
recommendations as well.
[0084] In some cases, these various occasions or events may fall
onto a predictable calendar schedule. These occasions or events may
be common holidays such as Valentine's Day (where a red color is
often appropriate), Saint Patrick's day (where traditionally at
least one item should be green), and so on, as well as various user
determined calendar settings. Here the server (100) and database
(102) can be configured to receive calendar information pertaining
to the calendar scheduling of the at least one event types, and the
characteristics of the clothing most suitable (or unsuitable) for
these event types. The server (100) and system can then use this
calendar information to make calendar occasion optimized fashion
recommendations as well. Thus the system may automatically remind
the user to wear a green wardrobe item on Saint Patrick's Day every
year, even if the user might otherwise forget.
[0085] Various other scoring algorithm methods are also possible.
FIG. 10 further shows that in some embodiments the scoring
algorithm can also integrate raw data from multiple sources,
including sources outside of the social network, do various types
of data mining, and in turn use this information to further guide
the scoring algorithm.
* * * * *