FIELD OF INVENTION
The present invention generally relates to a method and a system for providing contextual
keyword collective for communication events between a user and a contact. More particularly,
the present invention relates to a system and a method for creating and assigning a contextual
meaningful, short gist or snippet of keywords to one or more communication events happening
over one or more communication platforms between a user and his/her contact at a user device.
BACKGROUND OF INVENTION
[002] With the advent of technology in communication services through wireless devices, a
user of a wireless device is able to easily communicate with other users via a variety of modes of
communications. Not just via voice or SMS messaging service, various other communication
services, like e-mails, instant messaging, social network services are implemented in the wireless
communication devices. Addition of such services expanded on the degree, granularity and range
of the information that could be shared between the user and his/her contacts. Using a variety of
communication services by the user, on the same wireless device, presents the user with an
enormous data or information shared with the other contacts. In addition, using multiple
communication services on the same wireless device consume more power and device hardware
functionalities.
[003] Further, in situations where the enormous data that is being shared between the user over
the multiple platforms and his/her multiple contacts, the user usually tends to forget or give less
attention to the depth of contextual understanding for the communication that is being conducted
between two users over the multiple communication services. Also, the user may tend to forget
3
his/her last conversation with a particular contact, which might be important for future
conversations. For being up to date with the recent conversations, the user has to navigate
through the multiple conversations happening on each of the multiplicity of communicating
services, for each desired contact. This may be time-taking, and in a lot of situations, where the
user is immediately presented with a call or an SMS or an instant message from a particular
contact, the user may not have time to do such activity, or simply in other scenarios may not
even opt to do such activities because of lack of time or interest.
[004] Further, in order to be updated with the recent communication there is a requirement
from the user to access each of the communication platforms, individually and read through the
recent communication with the contact. This requirement of accessing multiple communication
platforms consumes unnecessary processing power and battery life of a wireless device that
could be used for other critical services. The processing capabilities and power source is limited
in a wireless device as compared to a standalone computer.
[005] Furthermore, there have been prior art technologies which have provided a system or an
application capable of analyzing plurality of messages of a user, assigning them with a label
describing a topic of the messages and providing the label to the user on an interface. Other
technologies provide systems for collecting and presenting historical communication data for a
mobile device. The information presented to a user of the mobile communication device is the
user contact information, related contact information, communication history and files shared
between the user and the selected contact. There are other methods provided in the present
technologies for extracting keywords from interactive text dialog and providing the extracted
keywords as a dialog summary display. Although, there are presently technologies which are
4
providing a user with an overview of the previous conversations, but they fail to provide a
contextual importance to the conversations and provide a weightage to the contextual meaning
of the conversations.
[006] Hence, there is no existing platform that enables viable bridge between various
communication platforms and combines ease of use with contextual awareness for recent
communication history.
OBJECTS OF INVENTION
[007] The present invention overcomes the above mentioned problems with conventional
digital platforms. The present invention provides a method and a system for providing contextual
summary as contextual keyword collective or contextual keywords phrase for the recent
communication of a user with each of the contacts in a user device.
[008] Therefore, it is an objective of the present invention to provide a system, a method and an
apparatus that provide an intelligent, short communication summary, to a user for each of his/her
contacts in a communication device.
[009] It is an objective of the present invention to provide a method and a system for creating
and presenting in form of an intelligent keywords phrase, customized topic for communication of
the user with his/her contacts based on communication history in a multi-platform
communication device.
5
[0010] It is also an objective of the present invention to create the intelligent keywords phrases
representing recent conversations with the user contacts, based on extracting contextual meaning
of the recent conversations and determining weightage for a contextual syntax in a sentence.
[0011] It is further an objective of the present invention to determine a group of contacts of the
user who are communicating over the same/ similar topic for influencing the context and
suggestions for a message or a call received by the user.
[0012] The foregoing and other objects, features and advantages of the invention will be
apparent from the following more particular descriptions of exemplary embodiments of the
invention.
SUMMARY OF INVENTION
[0013] The present invention provides a system for creating and providing an intelligent
contextual keyword phrase to a user using a user device, the system comprising:
the user device including:
a memory storing the application executable by one or more processors at
the user device;
a contact list of one or more contacts stored in the memory;
one or more communication applications stored in the memory and
executable by one or more processors at the user device for providing
communication platforms to the user and executing plurality of communication
events at the user device; and
6
the application including:
a capturing module for creating communication information that includes
information related to a plurality of communication events between the user and
each contact in the contact list and one or more communication events of the user
on one or more web-based servers, and the capturing module arranges the
communication information in chronological order for communication events
related to each contact in the contact list and the web-based servers;
a summarization module for summarizing the chronologically arranged
captured communication information into a summarized gist;
a context generation module for providing a contextual meaning to every
word included in the summarized gist to create a contextual summary;
a weightage module for providing a contextual weightage to each word in
the contextual summary; and
a matrix creating module for creating a confusion matrix that correlates
the contextual summary with one or more other contacts in the contact list based
on conversations between the user and the one or more other contacts in the
contact list that are related to the contextual summary; and
wherein the application creates the intelligent contextual keyword phrase for one or more
communication events between the user and each contact in the contact list, and one or more
communication events of the user on the one or more web-based servers, the intelligent
contextual keyword phrase includes the contextual summary along with its associated confusion
matrix.
7
[0014] The present invention provides a method of creating and providing an intelligent
contextual keyword phrase to a user using a user device, the intelligent contextual keyword
phrase being created and provided to the user by an application stored in memory of the user
device and executable by one or more processors at the user device, the method comprising:
creating, by a capturing module of the application, communication information
that includes information related to a plurality of communication events between
the user and each contact in a user contact list and one or more communication
events of the user on one or more web-based servers, the contact list being stored
at the user device of the user, and the plurality of communication events being
executed at the user device;
arranging, by the capturing module, the communication information in
chronological order for the plurality of communication events related to each
contact in the contact list and the web-based servers;
summarizing, by a summarization module of the application, the chronologically
arranged captured communication information into a summarized gist;
providing, by a context generation module of the application, a contextual
meaning to every word included in the summarized gist to create a contextual
summary;
providing, by a weightage module of the application, a contextual weightage to
each word in the contextual summary; and
creating, by a matrix creating module of the application, a confusion matrix that
correlates the contextual summary with one or more other contacts in the contact
8
list based on conversations between the user and the one or more other contacts in
the contact list that are related to the contextual summary; and
wherein the application creates the intelligent contextual keyword phrase for one or more
communication events between the user and each contact in the contact list, and one or more
communication events of the user on the one or more web-based servers, the intelligent
contextual keyword phrase includes the contextual summary along with its associated confusion
matrix.
[0015] The present invention provides a non-transitory computer readable recording medium
having recorded therein at least part of a software application executable, at least partly, by one
or more processors of a user device to perform a method, the user device includes:
the one or more processors;
memory storing at least part of operating system to facilitate, at least in part, operations
of the software application; and
wherein the method comprising:
creating, by a capturing module of the application, communication information
that includes information related to a plurality of communication events between
the user and each contact in a user contact list and one or more communication
events of the user on one or more web-based servers, the contact list being stored
at the user device of the user, and the plurality of communication events being
executed at the user device;
9
arranging, by the capturing module, the communication information in
chronological order for the plurality of communication events related to each
contact in the contact list and the web-based servers;
summarizing, by a summarization module of the application, the chronologically
arranged captured communication information into a summarized gist;
providing, by a context generation module of the application, a contextual
meaning to every word included in the summarized gist to create a contextual
summary;
providing, by a weightage module of the application, a contextual weightage to
each word in the contextual summary;
creating, by a matrix creating module of the application, a confusion matrix that
correlates the contextual summary with one or more other contacts in the contact
list based on conversations between the user and the one or more other contacts in
the contact list that are related to the contextual summary; and
prompting the intelligent contextual keyword phrases that includes the contextual
summary along with its associated confusion matrix, by the application, to the
user, at the user device, at one or more triggering events, and
wherein the application creates the intelligent contextual keyword phrase for one or more
communication events between the user and each contact in the contact list, and one or more
communication events of the user on the one or more web-based servers, the intelligent
contextual keyword phrase includes the contextual summary along with its associated confusion
matrix.
10
BRIEF DESCRIPTION OF DRAWINGS
[0016] FIG. 1 illustrates an exemplary environment including a system for generating intelligent
contextual keyword collective for a plurality of communication events in a multi-communication
platforms environment, in accordance with an embodiment of the present invention;
[0017] FIG. 2 illustrates an exemplary environment including an apparatus for generation of the
intelligent contextual keyword collective, in accordance with an embodiment of the present
invention; and
[0018] FIG. 3 illustrates a flowchart showing a method for generating the intelligent contextual
keyword collective for a plurality of communication events in a multi-communication platforms
environment, in accordance with an embodiment of the present invention.
DETAILED DESCRIPTION OF INVENTION
[0019] This patent describes the subject matter for patenting with specificity to meet statutory
requirements. However, the description itself is not intended to limit the scope of this patent. The
principles described herein may be embodied in many different forms.
[0020] Illustrative embodiments of the invention now will be described more fully hereinafter
with reference to the accompanying drawings, in which some, but not all embodiments of the
invention are shown. Indeed, the invention may be embodied in many different forms and should
not be construed as limited to the embodiments set forth herein; rather, these embodiments are
11
provided so that this disclosure will satisfy applicable legal requirements. Like numbers refer to
like elements throughout.
[0021] The present invention provides an apparatus, a system, and a method for creating and
assigning ‘contextual keyword collective’ or ‘contextual keyword phrases’ collectively
representing a plurality of communication events between a user and his/her contacts via a user
device, such as a smart phone. The present invention captures a plurality of communication
events of the user with various contacts over various communication channels, provides them
with chronological imprints, generates contextual fields for important words used in the
communication events, determines weightage for each word along with its context, where the
weightage depends on a number of factors, such as frequently occurring words, dictionary
meaning, chronological appearance, and hence creates an intelligent and contextual keyword
phrase which represents a short summary for all the communication events collectively, between
the user and each of his/her contacts in the user device. Therefore, for all the communication
events happening between the user and his/her one contact (as an example one contact), a
contextual keyword phrase is created and presented to the user at a time, which represents a short
summary collectively for all those communication events.
[0022] FIG. 1 illustrates an exemplary environment including the system for generating
intelligent contextual keyword collective for a plurality of communication events in a multicommunication platforms environment, in accordance with an embodiment of the present
invention. The system 100 for generating intelligent contextual keyword collective comprises a
user device 102 and an application 104 residing on the user device 102. The user device 102 may
12
include an electronic communication device such as and not limited to a computer, a laptop, a
smart phone, a tablet, an I-Phone, an I-Pad, and the like. The computer implemented system 100
comprises the application 104, which may also be referred to as ‘app’ throughout the detailed
description without deviating from the meaning and scope of the present invention. The
application 104 may be a mobile application or a web-based application or both that is run and is
accessible on the user device 102. A user operates the user device 102 for wirelessly
communicating with other users, or contacts stored in the user device 102.
[0023] The user device 102 further includes an Input/ Output module 106 which may include
and is not restricted to a keyboard, a touch pad, a mouse, a camera, a speaker, a microphone, a
display screen and the like, utilized by the user to send and/or receive messages or calls and
execute a communication event. The user device 102 has a memory 108 (and/or processors) for
storing, running and processing the application 104, and also stores other entities, such as
software programs to run the application 104. Further, the user device 102 also includes a contact
list 110 stored in the memory 108. The contact list 110 of the user may include information
related to a number of contacts of the user, where the information may include and is not
restricted to name, phone number, address, email address, contact IDs on one or more
communication platforms or applications, a picture, and the like.
[0024] The user communicates with one or more of his/her contacts via one or more
communication applications 112 stored in the memory 108 of the user device 102. The
communication application 112 is any communication channel or platform which the user may
utilize to contact his/her contacts through the user device 102. Therefore, the one or more
13
communication applications 112 may include and are not restricted to an SMS service, a voice
call service, an email application, an audio/video application, such as Skype™, an instant
messaging application, such as WhatsApp™ and the like which allows the user to send and/or
receive a message, whether a voice message or a text message, to and/or from a contact. Further,
the user device 102 may also include a calendar 114 that stores events for the user with time and
date stamps.
[0025] The user may get involved in a plurality of communication events where communication
or messages from various contacts in the contact list 110 are exchanged over various
communication channels 112. The application 104 silently captures such communication events
in the background. The captured data is stored on the user device, 102, itself. These captured
communication events with multiple messages are then arranged in a chronological order for
each contact of the user. The chronologically arranged communication events are then processed
as part of an information retrieval process in the background, where the application 104 executed
in the user device 104 generates a contextual and intelligent keyword collective for all the earlier
communication events of each contact of the user. ‘Keyword collective’ may also be
interexchangeably used with ‘keyword phrase’ hereinafter throughout the description of the
present invention, without deviating from the meaning and scope of the present invention.
[0026] A contextual keyword phrase, generated by the application 104, collectively represents
the plurality of communication events, corresponding to a particular contact, where the
communication events include exchanging of messages, whether voice or text, between a
particular contact and the user. It may be apparent to a person skilled in the art that the messages
14
that can be exchanged by the user, and captured by the application 104 may include and is not
restricted to a voice message, a text message, a multimedia message, an image, a GIF, and the
like, without deviating from the meaning and scope of the present invention.
[0027] As mentioned above, the application 104 silently captures the communication events for
each contact, using a capturing module 116. The capturing module 116 extracts all the messages
included in the communication events for each contact and store them locally in the memory
108. Further, the application 104 also includes an information crawler 118 that continuously
crawls the Internet, and collects information from publically available trusted sources of
information which are related to the communication events for each contact. Furthermore, the
application 104 also extracts information available in the events stored in the calendar 114,
which are either related to the contact and/or the communication events with that particular
contact. After gathering all the information related to the communication events for each user
using the capturing module 116 and the information crawler 118, the application 104 locally
stores the gathered information in the memory 108 of the device. In an embodiment, the
application 104 may also store the gathered information related to the communication events for
each user on a web-server, such as a cloud server, and may extract the information from the
server whenever needed.
[0028] The information, including the captured messages, information extracted from the
publically available trusted sources and the calendar events 114, related to the communication
events between a particular contact and the user may be hereinafter referred to as
‘communication information’ for that particular contact. Thereafter collecting the
15
communication information, the application 104 arranges the communication information, for
each contact, in a chronological order. The chronologically arranged communication
information is then processed as part of an information retrieval process in the background for
providing the user with a contextual summarized keyword phrase and a contextual suggestion as
a response to the recent conversations with a contact on a user interface 120 of the user device
102.
[0029] While processing the captured communication information, the application 104
implements an information retrieval module 122 for generating the intelligent contextual
keywords phrases for the communication information for each contact and presenting, on the
user interface 120, a ‘contextual intelligent response’ corresponding to the communication
information. The information retrieval module 122 includes a summarization module 124, a
context generation module 126, a weightage module 128 and a matrix creating module 130. The
summarization module 124 summarizes the messages exchanged from a contact of the user, into
lesser number of sentences, such as summarized to one or two sentences only. The messages that
are summarized may include and are not limited to SMS, emails, instant messages exchanged
over one or more communication applications, voice messages, multimedia, and the like.
Further, plurality of words included in the summarized sentences are assigned with context fields
which reflect meaning and importance of the words by determining entities related to those
words, where the entities may include and not restrict to names, places, things, concepts, time,
date, organizations, recognition, designation, and the like. The assigning of context fields to the
summarized sentences is executed by the context generation module 126.
16
[0030] Thereafter, the weightage module 128 provides a weightage to each word depending on a
number of factors, including and not restricting to frequency of usage, chronological appearances
in the communication information, dictionary meaning and the like. After assigning a weight to
each of the word in a conversation with a contact, a confusion matrix is calculated with each
word being a feature, by the matrix creating module 130. The confusion matrix determines a
group of contacts talking about the same or similar topics of the conversations. These sets of
contacts talking about similar topics may be hereinafter termed as "Contact Groups" throughout
the description of the present invention. The conversations with various contacts in a "Contact
Group" will influence setting up of the context of a communication event, such as an incoming
or outgoing call and will also impact automatically suggested responses for the communication
events, such as in messages.
[0031] Resultantly, the application 104 creates the intelligent contextual keyword phrase for
each contact that accounts for the weightage provided to each word used in the conversation with
a particular contact and their importance in the conversations with the other contacts falling in
the contact groups of that contact. Detailed functions of the application 104 and the information
retrieval module 122 are explained in the next FIG. 2.
[0032] FIG. 2 illustrates an exemplary environment including an apparatus for generation of the
intelligent contextual keyword collective, in accordance with an embodiment of the present
invention. With time, as the user utilizes the user device 102 to contact one or more contacts of
his/her contact list 110 via the one or more communication platforms 112, the apparatus 200
including the application 104 silently captures the conversations in such communication events
happening between the user and the contact(s) in the background, using the capturing module
17
116. As apparent to a person skilled in the art, that the user may contact or exchange
conversations with the one or more contacts in a variety of formats, such as including and not
restricted to an SMS message, an instant message from over the various communication
platforms 112, an email, a voice message, a multimedia, an image, or any other message
exchanged on a social network platform. The application 104 is able to capture the conversations
in all the formats aforementioned, and is able to recognize the messages on the social network,
from a particular contact, by using the information stored in the contact list, such as a social
network ID.
[0033] Along with the above conversations, the application 104 also surfs the Internet
continuously to crawl and gather information related to the conversations of the user with the
contact(s) and conversations of the user with corporate communities or organizations either over
the Internet or via any communication platform using the user device. The application 104
gathers relevant information from other sources like Internet. This information gathering is based
on two factors, first being the topics discussed with the contact over various other channels and
second being industry specific data for example if it is a call from a bank or a credit card agency
then the information gathered by the application 104 may include dues and account information
that were sent by various banks to the user over emails and SMS messages. Another example
includes if the call is from a hospital, then information gathered by the application 104 may
include the last prescription, appointments and test reports. Therefore, the application 104 may
also gather information about corporate communications that the user is in contact with. The
application 104 churns out the relevant information from the data stored on the user device or
web portals of other service providers and renders them to a user on the fly.
18
[0034] Further, the application 104 also extracts information from the events stored in the
calendar 114 which relates to the contact(s) or the user or the conversations with the contact(s),
individually or in combination. Such information, captured by the application 104, is collectively
referred to as ‘communication information’ which collectively reflects the plurality of
conversations or communication events between the user and the contact(s), information from
the publically trusted sources and the calendar 114 events. The captured communication
information is stored in the memory 108 of the device 102. Further, the application 104 now
arranges the captured communication information in chronological order for each contact.
[0035] The information retrieval module 122, thereafter, processes the chronologically arranged
captured communication information using its one or more components. FIG. 2 illustrates the
functions of the one or more components of the information retrieval module 122 for processing
the communication information. The summarization module 124 executes summarization 202 of
the captured communication information to convert the communication information into a short
summary including lesser number of sentences than included in the captured communication
information, such as maximum of one or two sentences providing a crisp gist of the captured
communication information. For the messages which are not into text format, such as voice,
image, multimedia, the application 104 may utilize artificial intelligent devices, (not shown in
the FIG. 2), such as voice recognition to convert voice into text, an image recognition for
converting image into a text describing the image, and a combination of both to convert a
multimedia message into plain text message. Therefore, the summarization module 124
summarizes the captured communication information into short gist (such as snippet). After
19
summarization 202, the application 104 executes clearing or removing 204 of redundant
unimportant non-contextual words referred to as ‘stop words’ from the summarized
conversations. Therefore, the application 104 cleans or removes the stop words, such as ‘is’, ‘a’,
‘an’, ‘the’, etc. from the summarized conversation.
[0036] Now the application 104 is presented with the summary which is further shortened, and
which is ready for context or meaning generation 206 of the summary. Therefore, the context
generation module 126 processes the summary of the captured communication information to
determine and provide a context field to the remaining words in the summary for creating a
contextual summary. In an embodiment, the application 104 determines the context field based
on NER method (Name, Entity, Recognition), in which the context recognizes and tells whether
the word is a person, place, location, time, day, organization, verb and the like. It may be
apparent to a person skilled in the art that the application 104 may implement other context
generation method present in the prior art, such as clustering and parsing based on syntax, and
including others, without deviating from the meaning and scope of the present invention.
[0037] The application 104 further provides weightage to each word in the contextual
summary, depending on the number of factors, using the weightage module 128. In an
embodiment, the weightage module 128 considers factors including and not limited to the
frequency of occurrences, chronological appearances, dictionary meaning, and the like to
calculate weightage of each word. The application 104 performs frequency analysis 208 on the
contextual summary generated by the context generation module 126. In frequency analysis 208,
executed by the weightage module 128, all the words in the contextual summary are counted to
20
determine their frequency of occurrences in all the communication events for a contact, also
using the context attached to the words. Further, the frequency count for each word, along with
its context, acts as an input for deciding on the weight of the word present in the contextual
summary. Along with the frequency analysis 208, the weightage module 128 also performs time
stamp analysis 210 for considering the chronological position of the communication events, in
which the word appears, to calculate the weightage of the word. Therefore, a word appearing in a
recent communication event is given more weightage. In addition to frequency and chronological
order analysis, the weightage module 128 also considers the dictionary importance 212 of the
word present in the contextual summary in order to decide the weightage of the word. For doing
this, application 108 may have access to a dictionary available on the Internet or locally stored in
the user device 102. Consequently, the weightage module 128 assigns weight to each word
appearing in the chronologically arranged contextual summary based on the factors, such as
including and not limited to frequency of occurrences 208, chronological order 210 and
dictionary importance 212, and the like.
[0038] Thereafter, the application 104 calculates a confusion matrix 214 using the matrix
creating module 130. After assigning weight to each of the word in the contextual summary
representing communication information of the user with a contact, and also with all the
contacts, a confusion matrix is calculated with each word being a feature. This confusion matrix
acts as an estimation of a correlation coefficient between the various conversations (or
communication events) of the user with his/her various contacts. These correlation coefficients
provide those contacts of the user who are talking about same/ similar topics and such contacts
are grouped together by the application 104. These sets of contacts, talking about the same/
21
similar topics, may be termed as "Contact Groups". The conversations with various contacts in a
"Contact Group" may influence determining of the context of an upcoming communication
event, such as an incoming or outgoing call and may also impact automatically suggested
responses for the communication event, such as the messages. Therefore, a user is prompted with
the conversation that he was having with the contact, while the application also tells the user that
which other contacts of the user are talking to him about similar topics, by showing him the
contacts in the ‘contact group’. Each contact is assigned various topics that the contact is talking
about with the user. Then a cross-correlation is calculated between each contact of the user and
the set of users that are talking about similar topics, have maximum correlation and can be
grouped into one contact group. Hence, creating a confusion matrix.
[0039] Further to creating contact groups, the application 104 creates intelligent keyword
phrases or keyword collectives 216 for each contact, which accounts for weighted words used in
the conversations in the communication events with that contact, and further to their importance
in the conversations with other contacts belonging to the contact group(s) of that contact.
Therefore, the keyword phrase or the keyword collective represents and highlights the
importance of all the conversations with a contact, in a short crisp summary, while also bringing
contextual meaning to the keyword phrase based on their frequency and time of occurrence and
dictionary importance; along with the information shared with other contacts on similar topics
and also the information available publically over the similar topics; and in addition, to the
related information stored in the calendar events.
22
[0040] The application 104 stores the keyword phrases for each contact in the memory 108 of
the user device 102. The keyword phrase is then presented (shown by 218) to the user on the user
interface 120 of the application 108, allowing the user to gain contextual awareness of the recent
conversation of the user with a contact. The keyword phrase is invoked and presented to the user,
by the application 104, on a number of triggering events. In an embodiment, the application 104
presents the keyword phrase to the user on two triggering events. In an embodiment, one
triggering event may be when the user chooses a contact from the contact list. In another
embodiment, the triggering event may be when the user receives or makes a phone call from or
to the contact. In yet another embodiment, the triggering event may be when the user receives or
sends a message, including an SMS or an instant message, or an email, or other formats, from or
to a contact. When such triggering events are detected by the application 104, the application 104
presents the generated keyword phrase to the user on the user interface 120.
[0041] In an exemplary environment, the application 104 promotes about the dues and account
information that were sent by various banks to the user over emails and SMS messages whenever
a triggering event like a call from a bank or a credit card agency is received by the user. In
another exemplary environment, the application 104 promotes last prescription, appointments or
test reports to the user whenever a triggering event like a call from a hospital is received by the
user. So it is not just one to one communication, but also about corporate communications and
information retrieval system that churns out the relevant information from the data stored on the
phone or web portals of other service providers and renders them to a user on the fly.
23
[0042] FIG. 3 illustrates a flowchart showing a method 300 for generating the intelligent,
contextual keyword collective for a plurality of communication events in a multi-communication
platforms environment, in accordance with an embodiment of the present invention. The method
is read considering the components of the FIGs. 1 and 2. The method 300 includes a step 302,
where the application 104 silently captures information or messages exchanged in plurality of
communication events between the user and a contact. The user utilizes the user device 102 for
communicating with one or more of the contacts over the multiple communication channels 112
stored in the user device 102. The messages exchanged in the plurality of communication events
between the user and a contact are captured by the application 104 in the background using the
capturing module 116. In an embodiment, the messages may be in a variety of formats, such as
including and not restricted to an SMS message, an instant message from over the various
communication platforms 112, an email, a voice message, a multimedia, an image, or any other
message exchanged on a social media platform. Along with gathering these messages, the
application 104, simultaneously, also crawls the Internet to collect and save the information
available publicly which is related to the communication events between the user and the
contact(s). In addition to this, the application 104 also extracts information stored in the calendar
events which are related to the communication events between the user and the contact(s). Such
information, captured by the application 104 at the step 302, is collectively referred to as
‘communication information’ which reflects the plurality of conversations or communication
events between the user and the contact(s), information from the publically trusted sources and
the calendar 114 events.
24
[0043] Further at a step 304, the application 104 arranges the communication information in
chronological order, for each contact. Thereafter, the application 104 implements the information
retrieval module 122 to process the chronologically arranged communication information. while
processing, the application 104 implements the summarization module 124, at step 306, for
summarizing all the communication events in the communication information into a short, crisp
summary, including and not restricted to one or two sentences. In an embodiment, the
application 104 may implement one or more artificial intelligent devices for the messages which
are not into text format, such as voice, image, multimedia, for example to convert voice into text,
an image recognition for converting image into a text describing the image, and a combination of
both to convert a multimedia message into plain text message. Therefore, the summarization
module 124 summarizes the captured messages into short snippet.
[0044] Thereafter at a step 308, the application 108 removes redundant unimportant words, from
the summarized snippet, which are referred to as ‘stop words’ from the summarized
conversation. Therefore, the application 104 cleans or removes the stop words, such as ‘is’, ‘a’,
‘an’, ‘the’, etc. at the step 308. After removing the stop words, the application 104, at step 310,
creates a context field for each word in the summarized snippet or gist, using the context
generation module 126. In an embodiment, the application 104 determines the context field
based on NER method (Name, Entity, Recognition), in which the context recognizes and tells
whether the word is a person, place, location, time, day, organization, verb and the like. It may
be apparent to a person skilled in the art that the application 104 may implement other context
generation method present in the prior art, such as clustering and parsing based on syntax, and
including others, without deviating from the meaning and scope of the present invention.
25
[0045] Further at a step 312, the application 104 implements the weightage module 128 for
assigning a weight to each word present in the contextual summary or the contextual summarized
snippet. The application 104 determines weight of each word in the contextual summary based
on a number of factors, including and not restricting to frequency analysis of each word,
chronological occurrence of each word, dictionary importance of each word and the like.
Therefore, the application 104 may first count the number of times each word is occurring in all
the communication events, along with considering its context in the events. Hence, for each word
in the contextual summary, a weight is calculated taking the frequency and the context as input.
In addition to this, the weight of the word is also affected by the chronological occurrence of the
communication event in which that word is appearing. Therefore, a word appearing in a recent
communication event is given more weightage. Further, the weight of the word is also affected
by its dictionary importance. For doing this, application 108 may have access to a dictionary
available on the Internet or locally stored in the user device 102. Consequently, the weightage
module 128 assigns weight to each word appearing in the chronologically arranged contextual
summary, at the step 312.
[0046] Furthermore at step 314, the application 104 implements the matrix creating module 130
for generating a confusion matrix. After the application 104 assigns weight to each word in the
contextual summary of communication events with a contact, and also across all the contacts, a
confusion matrix is created, which represents a plurality of contacts who are talking about the
similar topics with the user. such contacts are grouped as ‘contact groups’, and the conversations
with the contacts of the contact groups may influence determining of the context of an incoming
26
or outgoing call, and also may impact creating a suggested response for the messages from the
contacts.
[0047] Thereafter at step 316, the application 104 creates intelligent keyword phrases or
keyword collectives for each contact, where the keyword phrase represents the recent
conversations of the user with the contact, in a short contextual summary, bringing contextual
importance to the keyword phrase based on their frequency and time of occurrence and
dictionary importance; along with the information shared with other contacts on similar topics
and also the information available publically over the similar topics; and in addition, to the
information stored in the calendar events. Consequently, the application 104 presents the
keyword phrase to the user at the user interface 120, at step 318, whenever the application 104
detects a triggering event. In another embodiment, a triggering event may be when the user
chooses a contact from the contact list. In an embodiment, the triggering event may be when the
user receives or makes a phone call from or to the contact. In yet another embodiment, the
triggering event may be when the user receives or sends a message, including an SMS or an
instant message, or an email, or other formats, from or to a contact. When such triggering events
are detected by the application 104, the application 104 presents the keyword phrase to the user
on the user interface 120.
[0048] Advantageously, the present invention provides a system, a method and an apparatus for
creating a contextual keyword phrase(s) highlighting the essence of the recent communication
events of the user with his/her contact(s), which accounts for the weightage given to each word
in the keyword phrase depending on the frequency and chronological occurrences; and
27
contextual meanings of the words in the communication events, along with the dictionary
importance of the words. The keyword phrases are presented to the user on one or more
triggering events, such as while receiving a call, or a message from the contact.
[0049] Further, the objective of creating the Keyword Collective for each contact is to highlight
the important recent conversations with a contact and bring into the context, the information
shared on similar topics with other contacts and the information available in public over the
similar topics; also including the information stored in the calendar of the user. Such keyword
phrases make the user conversations more relevant as the user is already aware of the previous
communications in contextual manner. There is lesser requirement from the user to browse
through other communication platforms to come up to the same level of understanding with the
contact.
[0050] Furthermore, the present invention may also provide contextual suggestions, along with
the contextual summary, as a response to be provided by the user, for the recent conversations, in
any triggering event. For example, while writing an email to a contact, the application may
provide a contextual summary to the user highlighting the previous conversations; additionally,
the application may also provide a contextual suggestion which can be opted by the user to be
written as a response in the email.
[0051] Therefore, the present invention not only provides contextual summary of the previous
conversations, but also provide weightage to the words in the contextual summary while
formulating it, where the weightage may depend on number of times a word is occurring in the
28
previous communication events along with its context field; chronological order of the
communication events in which that word is appearing; and the dictionary importance of the
word. Along with this, the present invention also influences the context in the summary
depending on the contact groups sharing the similar topics, information available in public
sources over similar topics, and calendar events related to user, contacts, or similar topics.
[0052] While certain embodiments have been described, these embodiments have been presented
by way of example only, and are not intended to limit the scope of the present disclosure.
Indeed, the novel methods, devices, and systems described herein may be embodied in a variety
of other forms. Furthermore, various omissions, substitutions, and changes in the form of the
methods, devices, and systems described herein may be made without departing from the spirit
of the present disclosure.

We Claim:
1. A system for creating and providing an intelligent contextual keyword phrase to a user
using a user device, the system comprising:
the user device including:
a memory storing the application executable by one or more processors at
the user device;
a contact list of one or more contacts stored in the memory;
one or more communication applications stored in the memory and
executable by one or more processors at the user device for providing
communication platforms to the user and executing plurality of communication
events at the user device; and
the application including:
a capturing module for creating communication information that includes
information related to a plurality of communication events between the user and
each contact in the contact list and one or more communication events of the user
on one or more web-based servers, and the capturing module arranges the
communication information in chronological order for communication events
related to each contact in the contact list and the web-based servers;
a summarization module for summarizing the chronologically arranged
captured communication information into a summarized gist;
30
a context generation module for providing a contextual meaning to every
word included in the summarized gist to create a contextual summary;
a weightage module for providing a contextual weightage to each word in
the contextual summary; and
a matrix creating module for creating a confusion matrix that correlates
the contextual summary with one or more other contacts in the contact list based
on conversations between the user and the one or more other contacts in the
contact list that are related to the contextual summary; and
wherein the application creates the intelligent contextual keyword phrase for one or more
communication events between the user and each contact in the contact list, and one or more
communication events of the user on the one or more web-based servers, the intelligent
contextual keyword phrase includes the contextual summary along with its associated
confusion matrix.
2. The system as claimed in claim 1, wherein the communication information is created by
capturing conversations between the user and the one or more contacts of the contact list
and gathering information, related to the user, from the one or more web-based servers
and gathering information from one or more events stored in a calendar at the user
device.
3. The system as claimed in claim 1, wherein the summarization module further clears
redundant unimportant non-contextual words referred to as ‘stop words’ from the
summarized gist, the stop words include ‘is’, ‘a’, ‘an’, ‘the’ and other non-contextual
words.
31
4. The system as claimed in claim 1, wherein the context generation module determines the
contextual meaning of each word in the summarized gist based on an NER method
(Name, Entity, Recognition) in which the contextual meaning recognizes and tells
whether the word is a person, place, location, time, day, organization, or a verb.
5. The system as claimed in claim 1, wherein the context generation module implements
clustering and parsing methods based on syntax for providing contextual meaning to the
words in the summarized gist and creating the contextual summary.
6. The system as claimed in claim 1, wherein the weightage module provides the contextual
weightage to each word in the contextual summary by:
implementing frequency analysis that includes calculating frequency of
occurrences of each word, along with its contextual meaning, in the one or more
communication events; and
implementing time stamp analysis that includes determining chronological
appearances of each word in the one or more communication events; and
by determining dictionary meaning of each word.
7. The system as claimed in claim 1, wherein the matrix creating module correlates the
contextual summary with ‘contact groups’, the contact groups include the one or more
contacts in the contact list that communicate with the user on similar topics as that of the
contextual summary.
8. The system as claimed in claim 1, wherein the communication events include
conversation such as voice calls or exchanging SMS message, an instant message from
over the communication platforms, an email, a voice message, a multimedia, an image, or
any other message exchanged on a communication platform.
32
9. The system as claimed in claim 1, wherein the application further prompts the intelligent
contextual keyword phrases to the user, at the user device, at one or more triggering
events, the one or more triggering events include when the user chooses a contact from
the contact list, or when the user receives or makes a phone call from or to the contact, or
when the user receives or sends a message including an SMS or an instant message, or an
email, or other formats, from or to a contact.
10. The system as claimed in claim 1, wherein the user device is a computing device capable
of executing the plurality of communication events related to the user, the computing
device includes a laptop, a desktop, a smart phone, an I-phone, a tablet, or an I-Pad, and
the plurality of communication events are being executed over a network that includes
the Internet.
11. A method of creating and providing an intelligent contextual keyword phrase to a user
using a user device, the intelligent contextual keyword phrase being created and provided
to the user by an application stored in memory of the user device and executable by one
or more processors at the user device, the method comprising:
creating, by a capturing module of the application, communication information
that includes information related to a plurality of communication events between
the user and each contact in a user contact list and one or more communication
events of the user on one or more web-based servers, the contact list being stored
at the user device of the user, and the plurality of communication events being
executed at the user device;
33
arranging, by the capturing module, the communication information in
chronological order for the plurality of communication events related to each
contact in the contact list and the web-based servers;
summarizing, by a summarization module of the application, the chronologically
arranged captured communication information into a summarized gist;
providing, by a context generation module of the application, a contextual
meaning to every word included in the summarized gist to create a contextual
summary;
providing, by a weightage module of the application, a contextual weightage to
each word in the contextual summary; and
creating, by a matrix creating module of the application, a confusion matrix that
correlates the contextual summary with one or more other contacts in the contact
list based on conversations between the user and the one or more other contacts in
the contact list that are related to the contextual summary; and
wherein the application creates the intelligent contextual keyword phrase for one or more
communication events between the user and each contact in the contact list, and one or more
communication events of the user on the one or more web-based servers, the intelligent
contextual keyword phrase includes the contextual summary along with its associated confusion
matrix.
12. The method as claimed in claim 11 further comprises prompting the intelligent contextual
keyword phrases, by the application, to the user, at the user device, at one or more triggering
events, the one or more triggering events include when the user chooses a contact from the
contact list, or when the user receives or makes a phone call from or to the contact, or when the
34
user receives or sends a message including an SMS or an instant message, or an email, or other
formats, from or to a contact.
13. The method as claimed in claim 11, wherein the communication information is created by
capturing conversations between the user and the one or more contacts of the contact list and
gathering information, related to the user, from the one or more web-based servers and gathering
information from one or more events stored in a calendar at the user device.
14. The method as claimed in claim 11, wherein the summarization module further clears
redundant unimportant non-contextual words referred to as ‘stop words’ from the summarized
gist, the stop words include ‘is’, ‘a’, ‘an’, ‘the’ and other non-contextual words.
15. The method as claimed in claim 11, wherein the context generation module determines
the contextual meaning of each word in the summarized gist based on an NER method (Name,
Entity, Recognition) in which the contextual meaning recognizes and tells whether the word is a
person, place, location, time, day, organization, or a verb.
16. The method as claimed in claim 11, wherein the weightage module provides the
contextual weightage to each word in the contextual summary by:
implementing frequency analysis that includes calculating frequency of
occurrences of each word, along with its contextual meaning, in the one or more
communication events; and
implementing time stamp analysis that includes determining chronological
appearances of each word in the one or more communication events; and
by determining dictionary meaning of each word.
17. The method as claimed in claim 11, wherein the matrix creating module correlates the
contextual summary with ‘contact groups’, the contact groups include the one or more contacts
35
in the contact list that communicate with the user on similar topics as that of the contextual
summary.
18. The method as claimed in claim 11, wherein the communication events include
conversation such as voice calls or exchanging SMS message, an instant message from over the
communication platforms, an email, a voice message, a multimedia, an image, or any other
message exchanged on a communication platform.