TECHNICAL FIELD
Embodiments of the present invention generally relate to the field of mobile communication. More particularly, to methods and systems for creating a sticker store and determining sticker response for recommendation.
BACKGROUND 5
This section is intended to provide information relating to general state of the art and thus any approach/functionality described below should not be assumed to be qualified as a prior art merely by its inclusion in this section.
In today’s world, communication of information is spearheading to a new level of advancement with the advent of systems facilitating communication between two 10 or more users. Largely, the communication between the two or more users includes activities, for example, creating, storing, exchanging and managing electronic messages in the form of text, audio, video, image, or a combination thereof. Within these, with the rapid increase in the use of the electronic messaging for communication, there has emerged a need for developing systems and methods 15 that make the electronic messaging process easier, faster and more efficient. More particularly, there is a need for developing systems that aids the user in replying to a message. In view of this, several messaging solutions have been developed that help a user to reply to the message in a better, faster or more efficient manner. However, such existing solutions have a number of limitations and drawbacks. The existing 20 solutions primarily use text as means of communication. Text messages in the electronic messaging sphere are incapable of conveying different forms and degrees of expressions effectively. Thus, other forms of visual expressions such as stickers are used in messaging to enable effective conveying of users’s expressions. Further,
3
sending text messages is often time consuming, especially when an illustration may easily convey the message.
Certain existing solutions provision features such as spell correction, text prediction and sending visual expressions or ideograms. While sending the visual expressions or ideograms in response to the message, to some extent, helps to identify the 5 sender’s intent, however, the sender is provided only with a limited number of options which often again makes it difficult to indicate the appropriate intent. Further, selecting the visual expressions or ideograms to be sent in response to a message requires the sender to scroll through an entire list of available options to find the one closest to his intent which is a cumbersome and a time-consuming task 10 and thus, is generally not preferred by the users.
Therefore, in view of the above drawbacks and limitations of the existing solutions, there is a need to develop a system and method that recommends or suggests appropriate replies to the user in response of the message received from another user. 15
SUMMARY
This section is provided to introduce certain objects and aspects of the present disclosure in a simplified form that are further described below in the detailed description. This summary is not intended to identify the key features or the scope of the claimed subject matter. 20
Embodiments of the present invention may relate to a method comprising: analyzing a social-network data to determine one or more response stickers corresponding to a sticker, wherein said social-network data includes at least a chat history of one or more users; storing a first correlation between the sticker and the one or more received response stickers at a sticker suggestion store, wherein said 25
4
first correlation is automatically assigned a first confidence score, wherein the first confidence score is dynamically updated based on the analysis; receiving a new sticker at the sticker suggestion store; based on the stored first correlation, determining a response sticker set correlated to the received new sticker, wherein a second correlation between the new sticker and the response sticker set is 5 automatically assigned a second confidence score; and based on the first confidence score and the second confidence score, recommending, in response to one of the sticker and the new sticker, at least one of said one or more received response stickers, said response sticker set and a combination thereof.
Embodiments of the present invention may further relate to a sticker 10 recommendation system comprising: an analysis module configured to analyze a social-network data to determine one or more response stickers corresponding to a sticker, wherein said social-network data includes at least a chat history of one or more users; a sticker suggestion store configured to store a first correlation between the sticker and the one or more response stickers, based on the analysis, and receive 15 a new sticker; a recommendation module configured to determine a response sticker set correlated to the received new sticker based on the stored first correlation, and recommend, in response to one of the sticker and the new sticker, at least one of said one or more received response stickers, said response sticker set and a combination thereof; and a training module configured to automatically assign 20 a first confidence score to said first correlation, wherein the first confidence score is dynamically updated based on the analysis, and automatically assign a second confidence score to a second correlation, between the new sticker and the response sticker set, wherein the recommendation is based on the first confidence score and the second confidence score. 25
5
BRIEF DESCRIPTION OF THE DRAWINGS
The accompanying drawings, which are incorporated herein, and constitute a part of this disclosure, illustrate exemplary embodiments of the disclosed methods and systems in which like reference numerals refer to the same parts throughout the different drawings. Components in the drawings are not necessarily to scale, 5 emphasis instead being placed upon clearly illustrating the principles of the present disclosure. Some drawings may indicate the components using block diagrams and may not represent the internal circuitry of each component. It will be appreciated by those skilled in the art that disclosure of such drawings includes disclosure of electrical components or circuitry commonly used to implement such components. 10
FIG.1 illustrates an overall communication between a sticker recommendation system [100] and at least one user device [104A, 104B, 104C], in accordance with an embodiment of the present invention.
FIG.2 illustrates an exemplary sticker recommendation system [100], in accordance with an embodiment of the present invention. 15
FIG.3 illustrates a method flow diagram [300] for creating a sticker suggestion store [208], in accordance with an embodiment of the present invention.
FIG.4 illustrates a method flow diagram [400] for recommending at least one or more response stickers, a response sticker set and a combination thereof, in accordance with an embodiment of the present invention. 20
FIG.5 illustrates a method flow diagram [500] for recommending a sticker, in accordance with an embodiment of the present invention.
6
FIG.6 illustrates an exemplary screenshot [600] displaying sticker recommendation on at least one user device [104A, 104B, 104C], in accordance with an embodiment of the present invention.
FIG.7 illustrates an exemplary screenshot [700] displaying sticker recommendation on at least one user device [104A, 104B, 104C], in accordance with an embodiment 5 of the present invention.
DETAILED DESCRIPTION
In the following description, for the purposes of explanation, various specific details are set forth in order to provide a thorough understanding of embodiments of the present invention. It will be apparent, however, that embodiments of the present 10 invention may be practiced without these specific details or with additional details that may be obvious to a person skilled in the art. Several features described hereafter can each be used independently of one another or with any combination of other features. An individual feature may not address any of the problems discussed above or might address only one of the problems discussed above. Some 15 of the problems discussed above might not be fully addressed by any of the features described herein. Example embodiments of the present invention are described below, as illustrated in various drawings in which like reference numerals refer to the same parts throughout the different drawings.
As used herein, a sticker may denote a graphic image used during a chat, on a 20 messaging application or a social network service, for expressing an emotion or an action through cartoons, animations and emojis. Further, the sticker may be used interchangeably with an emoticon.
As used herein, a sticker recommendation system [100] refers to a recommendation system that may create or introduce new stickers, store already existing one or 25
7
more stickers and one or more response stickers. Further, the sticker recommendation system [100] tracks chat history of one or more users of the messaging application or the social network service to recommend one or more response stickers to the users. Based on the tracked chat history, the sticker recommendation system [100] identify a correlation between the one or more 5 stickers and the one or more response stickers. Also, the sticker recommendation system [100] provide a confidence score to the identified correlation between the one or more stickers and the one or more response stickers. The response stickers may be a reply sticker or a follow-up sticker. In particular, if a first user sends a first sticker to a second user during a chat and the second user sends a second sticker as 10 a reply to the first user, then the second sticker sent as a reply is considered as the “reply sticker”. Further, if the second user does not reply back, then the first user sends a third sticker to the second user, then the third sticker sent to the second user is considered as the “follow-up sticker”.
As used herein, a sticker suggestion store in the sticker recommendation system 15 [100] refers to any database that stores the one or more stickers, a new sticker and the one or more response stickers. In specific, the one or more stickers are the previously stored stickers which already exist at the sticker suggestion store [208].
As used herein, a new sticker refers to any sticker that is not previously stored on sticker suggestion store [100]. 20
As used herein, the correlation may define a relationship between one or more stickers. In particular, the correlation refers to which stickers can be used as a response to a specific sticker.
8
As used herein, a first confidence score is a numeric value that may be assigned to the correlation between the one or more stickers and the one or more response stickers. The first confidence score may range from 0 to any number. This score is identified based on the analysis of social network data including chat history. Similarly, a second confidence score is a numeric value that may be assigned to the 5 correlation between the new stickers and the one or more response stickers of the new sticker. The second confidence score may range from 0 to any number. This score is also identified based on the tracked chat history. The details of how the confidence score is assigned is provided hereinbelow.
Embodiments of the present invention relates to systems and methods for creating 10 and maintaining the sticker recommendation system [100] by storing a correlation between one or more stickers and its corresponding one or more response stickers. The correlation between the one or more stickers and its corresponding one or more response stickers is assigned a first confidence score based on an analysis of a social network data. 15
Further, the invention also encompasses a method and a system for recommending one of the one or more stickers and the one or more response stickers, from the sticker recommendation system [100], to at least one of the user device [104A, 104B, 104C] based on the first confidence score.
Further, the invention further encompasses a method and a system for 20 recommending a new sticker, from the sticker recommendation system [100] to at least a subset of user device [104A, 104B, 104C] and assigning a second confidence score to a second correlation stored in the sticker recommendation system [100], between the new sticker and a response sticker set.
25
9
The response stickers as used herein may include, but are not limited to, a reply sticker, a follow-up sticker and any such sticker that may be recommended to the user obvious to a person skilled in the art that may be suited for a response.
The social network data as used herein may include, but are not limited to, a chat history data of at least one user/at least one user device [104A, 104B, 104C], a usage 5 history of any sticker, and any such social network data obvious to a person skilled in the art.
The at least one user device [104A, 104B, 104C] as used herein may include, but are not limited to, a mobile phone, a tablet, a phablet, a laptop, a desktop computer, a personal digital assistant (PDA), a plain old telephone service device and any such 10 device obvious to a person skilled in the art. Further, a user is associated with the at least one user device [104A, 104B, 104C] and the at least one user device [104A, 104B, 104C] comprises an operating system, a memory unit, a processor, a display interface, an input means such as a keyboard or touch input etc.
Exemplary Table 1 illustrates the one or more stickers (Sticker A, Sticker B, Sticker C, 15 Sticker D & Sticker “Hi”). Each sticker has the one or more response stickers (i.e. reply sticker, follow-up sticker). As mentioned, the Sticker A has three reply stickers namely SAR1, SAR2, SAR3 and three follow-up stickers i.e. SAF1, SAF2, SAF3. Sticker B has three reply stickers, SBR1, SBR2, SBR3 and two follow-up stickers, SBF1, SBF2. Similarly, Sticker C has two reply stickers, SCR1, SCR2 and four follow-up stickers 20 (SCF1, SCF2, SCF3, SCF4). Sticker D has four reply sticker SDR1, SDR2, SDR3, SRD4 and three follow-up stickers, SDF1, SDF2, SDF3. Lastly, “Hi” sticker has "Hello", “Hey”, “Hiiiii” as reply stickers and "Whatsup", "Everything ok", "Are you alive" as follow-up stickers.
25
10
Sticker
Reply Sticker
Follow-up Sticker
Sticker A
SAR1, SAR2, SAR3
SAF1, SAF2, SAF3
Sticker B
SBR1, SBR2, SBR3
SBF1, SBF2
Sticker C
SCR1, SCR2
SCF1, SCF2, SCF3, SCF4
Sticker D
SDR1, SDR2, SDR3, SRD4
SDF1, SDF2, SDF3
Hi Sticker
"Hello", “Hey”, “Hiiiii”
"Whatsup", "Everything ok", "Are you alive"
Table 1
As illustrated in FIG.1, the present invention illustrates an overall communication between a sticker recommendation system [100] and at least one user device [104A, 104B, 104C], in accordance with an embodiment of the present invention. 5
A first user associated with user device [104A] selects and sends at least one sticker to a second user associated with user device [104B] through a network [102]. In response to receiving the at least one sticker, the sticker recommendation system [100] determines one or more response stickers and recommends the one or more response stickers to the second user associated with user device [104B]. Then, the 10 second user selects and sends the at least one response sticker, from the recommended one or more response stickers, to the first user associated with user device [104A]. In another embodiment, the sticker recommendation system [100] determines and recommends the one or more response stickers to the first user associated with user device [104A]. Further, the communication between the first 15 user, the second user and the sticker recommendation system [100] takes place through the network [102]. Considering the example from the Table 1, the first user
11
sends the “Hi” sticker to the second user, the sticker recommendation system [100] determines and recommends "Hello", “Hey”, “Hiiiii” sticker/s to the second user as reply sticker/s. In case a reply for the “Hi” sticker is not received from the second user, then follow-up sticker/s such as "Whatsup", "Everything ok", "Are you alive" etc. are recommended, by the sticker recommendation system [100], to the first 5 user. Further, the follow-up sticker/s are recommended and displayed to the first user after a pre-determined time period when the reply for the “Hi” sticker is not received from the second user.
The specifics of how the sticker recommendation system [100] determines and recommends the one or more response stickers to the at least one user device 10 [104A, 104B, 104C], is described in Figure. 2.
As illustrated in FIG.2, the present invention illustrates the sticker recommendation system [100], in accordance with an embodiment of the present invention. The sticker recommendation system [100] comprises of, but not limited to, a user logs database [202], a recommendation module [204], an analysis module [206], a sticker 15 suggestion store [208] and a training module [210].
The user logs database [202] is configured to continuously track a social-network data of at least one user device [104A, 104B, 104C] wherein the social-network data includes a chat history of at least one user associated with at least one user device [104A, 104B, 104C]. The tracking of the social-network data includes, but to limited 20 to, tracking which response sticker is used as a reply and/or a follow-up to a sticker, which response sticker is viewed as the reply and/or the follow-up to the sticker but not selected as the reply and/or the follow-up, which response sticker is viewed and selected as the reply and/or the follow-up to the sticker and a combination thereof. For an instance, the user logs database [202] tracks that "Hey" sticker is the most 25
12
frequently viewed and selected sticker by one or more users as the reply sticker instead of “Hello” or “Hiiiii” sticker/s. Then, “Hello” sticker is viewed and selected sticker by the one or more users and lastly, “Hiiiii” sticker is the least frequently viewed and selected sticker by the one or more users. And, in case the reply for the “Hi” sticker is not received from the second user, then "Whatsup" sticker as the 5 follow-up sticker/s is most the frequently viewed and selected by the one or more users. The, "Are you alive" sticker is frequently viewed and selected by the one or more users and lastly, "Everything ok" is the least frequently viewed and selected by the one or more users. In an embodiment, the tracking of each sticker in the sticker recommendation system [100] is performed. In another embodiment, the tracking 10 of some sticker in the sticker recommendation system [100] is performed. The analysis module [206] is configured to use tracked social-network data, from the user logs database [202], as an input and based on the tracked social-network data, the analysis module [206] is further configured to determine a correlation between the one or more stickers and the one or more response stickers. Following the above 15 example, the analysis module [206] determines the correlation between the “Hi” sticker sent by the first user to the second user and "Hey", “Hello” and “Hiiiii” sticker sent as reply sticker/s sent from the second user to the first user. Similarly, a correlation between “Hi” sticker and "Whatsup”, "Are you alive" and "Everything ok" sticker is determined. This correlation between the “Hi” sticker and "Hey", “Hello” 20 and “Hiiiii” sticker sent as reply sticker/s is determined using the tracked social-network data. This means, that most of the users in the messaging application prefers using "Hey", “Hello” and “Hiiiii” stickers in reply to the “Hi” sticker. Thus, three different correlations for reply stickers between “Hi” - "Hey", “Hi”- “Hello” and “Hi”- “Hiiiii” are established. Similarly, a correlation between the “Hi” sticker and 25 “Whatsup”, "Are you alive" and "Everything ok" sent as follow-up sticker/s is determined using the tracked social-network data. This means, that most of the
13
users in the messaging application prefers using "Whatsup”, "Are you alive" and "Everything ok" stickers as follow-up to the “Hi” sticker. Thus, three different correlations for follow-up stickers between “Hi” - "Whatsup”, “Hi”- “Are you alive” and “Hi”- “Everything ok” are established.
In an embodiment, the analysed social-network data is the data of all the users of 5 the messaging application. Further, in another embodiment, the social-network data is analysed to determine a sequence of stickers exchanged during the chat in the messaging application.
Embodiments of the analysis module [206] further encompasses to compute the one or more response stickers for every sticker by mining chats between users of 10 the messaging application. In such embodiment, the analysis module [206] determines ordered pairs of stickers (x, y), where the sticker y has been shared as a response (a reply or a follow-up) to sticker x, in one or more chats between users. Considering two discrete random variables X and Y, where X and Y can take up all possible stickers x’s and y’s respectively such that (x, y) is an ordered pair of stickers 15 determined as above. Then, the analysis module [206] measures the probability, P (X=x, Y=y) across all observations (x, y). An embodiment, for creating response stickers for sticker x, selects any sticker y such that P (X=x, Y=y) is high. However, this approach may not yield relevant response stickers at all times. Consider the case where y is a highly popular sticker that is applicable in different contexts. For 20 instance, a sticker h with a happy face or a smile is often used. So, P (X=x, Y=h) is high for any choice of X. However, there could be more appropriate responses for stickers than a generic sticker with a happy face. In another embodiment, the analysis module [206] computes a confidence score as P (X=x, Y=y) / P (Y=y). Such a formulation discounts the confidence score for responses that are generic, i.e. 25 responses that are highly probable irrespective of the context. Yet another
14
embodiment, computes the confidence score with the help of pointwise mutual information (PMI). Further, the confidence score for y as a response for x is taken to be pmi (x;y) = log [P(X=x, Y=y) / (P(X=x) P(Y=y))]. In yet another embodiment, the analysis module [206] may use the normalized version of PMI.
The training module [210] is initialized by the analysis module [206] and is 5 configured to automatically assign a first confidence score to the correlation between the one or more stickers and the one or more response stickers wherein the training module [210] automatically assigns the first confidence score based on the tracked social-network data. Continuing with the above example, the training module [210] automatically assigns a confidence score of 10 to the correlation 10 between the “Hi” sticker and the "Hey" sticker which is the most frequently sticker sent as reply sticker. Further, a confidence score of 5 is assigned to a correlation between the "Hi" sticker and the “Hello” sticker and a confidence score of 2 is assigned, by the training module [210], to a correlation between the "Hi" sticker and “Hiiiii” sticker/s. The confidence scores of 5 and 3, respectively, are assigned to the 15 “Hello” and “Hiiiii” sticker/s as these are comparatively less frequently selected as reply to the “Hi” sticker. Similarly, the training module [210] automatically assigns a confidence score of 10 to the correlation between “Hi” sticker and "Whatsup" sticker. Also, a confidence score of 5 is automatically assigned to a correlation between the "Hi" sticker and the "Are you alive" sticker and a confidence score of 2 20 is assigned, by the training module [210], to a correlation between the "Hi" sticker and the "Everything ok"/sticker. The confidence scores of 5 and 2, respectively, are assigned to the "Are you alive" and "Everything ok" stickers, as these are comparatively less frequently selected as the follow-up sticker to the “Hi” sticker.
25
15
The present invention further encompasses the training module [210] configured to dynamically update the first confidence score assigned to the correlation between the one or more stickers and the one or more response stickers based on the continuously tracking of the social-network data. With the above example, the training module [210] automatically and dynamically update the confidence score 5 assigned to the correlation between the “Hi” sticker and the "Hey" sticker if, based on the analysis of the social-network data by the analysis module [206], it is determined that the "Hey" sticker is not anymore the most frequently sticker sent as reply sticker to the “Hi” sticker and now, the “Hello” sticker is the most frequently sent as reply sticker to the “Hi” sticker. In this scenario, the confidence score 10 assigned to the correlation between the "Hi" sticker and the “Hello” sticker is updated from 5 to 10. And, the confidence score assigned to the correlation between the "Hi" sticker and the “Hey” sticker is updated from 10 to 5.
Similarly, the training module [210] automatically update the confidence score from 10 to 5 assigned to the correlation between “Hi” sticker and "Whatsup" sticker, if it 15 is determined that the "Whatsup" sticker is not anymore the most frequently sticker sent as follow-up sticker to the “Hi” sticker based on the analysis of the social-network data by the analysis module [206]. And further, if the "Are you alive" sticker is the most frequently sent as follow sticker to the “Hi” sticker, the training module [210] automatically update the confidence score from 5 to 10 to the correlation 20 between “Hi” sticker and "Are you alive" sticker.
The recommendation module [204] is configured to determine a response sticker set based on the confidence score and recommend one of the response sticker set, the one or more response stickers, and a combination thereof, to the at least one user device [104A, 104B, 104C]. In an embodiment, the response sticker set may or 25 may not include the one or more response stickers. With the same example, the
16
reply sticker/s with high confidence score is determined as response sticker set. In the above case, the "Hey" sticker and the “Hello” sticker with the confidence scores of 10 and 5, respectively, is determined as the response sticker set and recommended as the reply sticker/s for the “Hi” sticker to the at least one user device [104A, 104B, 104C] by the recommendation module [204]. Also, "Whatsup" 5 sticker and the "Are you alive" sticker with the confidence scores of 10 and 5, respectively, is determined as the response sticker set and recommended as the follow-up sticker/s for the “Hi” sticker to the at least one user device [104A, 104B, 104C] by the recommendation module [204].
The sticker suggestion store [208] is configured to create and/or receive one or 10 more new sticker as well as store the new sticker and the at least one sticker. The one or more new sticker is either created/introduced at the sticker suggestion store [208] or received from a user. Moreover, the sticker suggestion store [208] is configured to store a plurality of the sticker, the new sticker and the response stickers. The one or more stickers are the sticker which already exist at the sticker 15 suggestion store [208] and are used by the at least one user device [104A, 104B, 104C]. The one or more stickers are stored at the sticker suggestion store [208] with its corresponding one or more response stickers. The sticker suggestion store [208] is further configured to store the correlation between the one or more stickers and the one or more response stickers. 20
FIG.3 illustrates a method flow diagram [300] for creating the sticker suggestion store [208], in accordance with an embodiment of the present invention.
The following includes detailed steps involved in creating the sticker suggestion store [208], wherein the method step initiates at step 302.
17
At step 304, a continuously tracking of the social-network data of the at least one user device [104A, 104B, 104C] is performed wherein the social-network data includes the chat history of at least one user device [104A, 104B, 104C]. Based on the tracked social-network data, the correlation is determined between the one or more stickers and the one or more response stickers. The tracking of the social-5 network data includes, but to limited to, tracking which response sticker is frequently used as a reply and/or a follow-up to a sticker, which response sticker is viewed as the reply and/or the follow-up to the sticker but not selected as the reply and/or the follow-up, which response sticker is viewed and selected as the reply and/or the follow-up to the sticker and a combination thereof. For an instance, the 10 user logs database [202] tracks that "Hey" sticker is the most frequently viewed and selected sticker by one or more users as the reply sticker instead of “Hello” or “Hiiiii” sticker/s. Then, “Hello” sticker is viewed and selected sticker by the one or more users and lastly, “Hiiiii” sticker is the least frequently viewed and selected sticker by the one or more users. And, in case the reply for the “Hi” sticker is not received from 15 the second user, then "Whatsup" sticker as the follow-up sticker/s is most the frequently viewed and selected by the one or more users. The, "Are you alive" sticker is frequently viewed and selected by the one or more users and lastly, "Everything ok" is the least frequently viewed and selected by the one or more users. Further, the correlation between the “Hi” sticker and "Hey", “Hello” and 20 “Hiiiii” sticker sent as reply sticker/s is determined using the tracked social-network data. This means, that most of the users in the messaging application prefers using "Hey", “Hello” and “Hiiiii” stickers in reply to the “Hi” sticker. Thus, three different correlations for reply stickers between “Hi” - "Hey", “Hi”- “Hello” and “Hi”- “Hiiiii” are established. Similarly, a correlation between the “Hi” sticker and “Whatsup”, 25 "Are you alive" and "Everything ok" sent as follow-up sticker/s is determined using the tracked social-network data. This means, that most of the users in the messaging
18
application prefers using "Whatsup”, "Are you alive" and "Everything ok" stickers as follow-up to the “Hi” sticker. Thus, three different correlations for follow-up stickers between “Hi” - "Whatsup”, “Hi”- “Are you alive” and “Hi”- “Everything ok” are established.
At step 306, the correlation between the one or more stickers and the one or more 5 response stickers is automatically assigned the first confidence score. Then, the first confidence score for the correlation between the one or more stickers and the one or more response stickers is stored. Continuing with the above example, automatically assigning the confidence score of 10 to the correlation between the “Hi” sticker and the "Hey" sticker most frequently sent as reply sticker. Further, the 10 confidence score of 5 to a correlation between the "Hi" sticker and the “Hello” sticker and a confidence score of 2 is assigned, by the training module [210], to a correlation between the "Hi" sticker and “Hiiiii” sticker/s. The confidence scores of 5 and 3, respectively, are assigned to the “Hello” and “Hiiiii” sticker/s as these are comparatively less frequently selected as reply to the “Hi” sticker. Similarly, 15 automatically assigning the confidence score of 10 to the correlation between “Hi” sticker and "Whatsup" sticker. Also, the confidence score of 5 is automatically assigned to the correlation between the "Hi" sticker and the "Are you alive" sticker and the confidence score of 2 is assigned to the correlation between the "Hi" sticker and the "Everything ok"/sticker. The confidence scores of 5 and 2, respectively, are 20 assigned to the "Are you alive" and "Everything ok" stickers, as these are comparatively less frequently selected as the follow-up sticker to the “Hi” sticker.At step 308, based on the first confidence score and the tracked social-network data, the response sticker set is determined and one of the response sticker set, the one or more response stickers, and a combination thereof is recommended to the at 25 least one user device [104A, 104B, 104C]. In an embodiment, the response sticker
19
with high confidence score is recommended as the response sticker set to the at least one user device [104A, 104B, 104C]. In the example, "Hey" sticker with the confidence score of 10 recommended as the reply sticker. Similarly, the "Whatsup" sticker with the confidence score of 10 recommended as the follow-up sticker. In an alternative embodiment, the response sticker with low confidence score may also 5 recommended as the response sticker set to the at least one user device [104A, 104B, 104C].
At step 310, the method flow [300] ends here.
FIG.4 illustrates a method flow diagram [400] for recommending one or more response stickers, a response sticker set and a combination thereof, in accordance 10 with an embodiment of the present invention.
The following includes detailed steps involved recommending the one or more response stickers, the response sticker set and the combination thereof, wherein the method step initiates at step 402.
At step 404, continuously tracking the social-network data of the at least one user 15 device [104A, 104B, 104C] wherein the social-network data includes the chat history of at least one user device [104A, 104B, 104C]. Based on the tracked social-network data, a correlation is determined between the one or more stickers and the one or more response stickers.
At step 406, the correlation between one or more stickers and the one or more 20 response stickers is automatically assigned the first confidence score. Then, the first confidence score for the correlation between the one or more stickers and the one or more response stickers is stored at the sticker suggestion store [208].
20
At step 408, a new sticker is received at the sticker suggestion store [208] wherein the new sticker is not previously stored at the sticker suggestion store [208]. For instance, an “ola” sticker is the new sticker received at the sticker suggestion store [208].
At step 410, on receiving the new sticker, a metadata of the new sticker is compared 5 with a metadata of the one or more stickers already stored at the sticker suggestion store [208]. The comparison of the metadata is performed in order to identify a similar sticker. The similar sticker has a metadata that is identical or similar matches with the metadata of the new sticker. Further, the similar sticker has associated one or more response stickers. After identifying the similar sticker, the one or more 10 response stickers of the similar sticker is now correlated with the new sticker. Then, a correlation between the new sticker and the one or more response stickers of the similar sticker is stored at the sticker suggestion store [208]. A second confidence score is automatically assigned to the correlation between the new sticker and the one or more response stickers by the training module [210]. The one or more 15 response stickers correlated to the new sticker is now called a response sticker set of the new sticker. Following the same example, the metadata of the “ola” sticker is compared with the metadata of the “Hi” sticker, “Hello” sticker already stored at the sticker suggestion store [208] with their corresponding one or more response stickers. Based on the comparison of the metadata, the “Hi” sticker is determined as 20 the similar sticker having the closet identical or similar match of the metadata. Once the “Hi” sticker is determined as the similar sticker, one or more response stickers of the “Hi” sticker now correlated with the “ola” sticker. In particular, "Hello", “Hey”, “Hiiiii” sticker/s are correlated as the reply stickers to the “ola” sticker and the "Whatsup", "Everything ok", "Are you alive" sticker/s are correlated as the follow-up 25 sticker/s to the “ola” sticker. Now, the "Hello", “Hey”, “Hiiiii” sticker/s as the reply
21
stickers and the "Whatsup", "Everything ok", "Are you alive" sticker/s as the follow-up sticker/s is determined as the response sticker set for the “ola” sticker based on the above metadata comparison.
The metadata, as used herein may include, but are not limited to, an identifier associated with a sticker, a particular group/theme associated with a sticker, a time 5 or a duration associated with a sticker, a sticker usage, and a popularity associated with a sticker, sentiment associated with a sticker and any such metadata obvious to a person skilled in the art.
At step 412, once the response sticker set for the new sticker is determined, one of the one or more response stickers, the response sticker set and a combination 10 thereof is recommended to the at least one user device [104A, 104B, 104C]. This recommendation is based on the first confidence score assigned to the correlation between the one or more stickers and the one or more response stickers and the second confidence score assigned to the correlation between the new sticker and the one or more response stickers. Continuing with the same example, the "Hello", 15 “Hey”, “Hiiiii” sticker/s as the reply stickers and the "Whatsup", "Everything ok", "Are you alive" sticker/s as the follow-up sticker/s for the “ola” sticker is recommended as the response sticker set to the at least one user device [104A, 104B, 104C]. Along with the response sticker set, the one or more response stickers of the one or more stickers stored at the sticker suggestion store [208] or the 20 combination thereof is recommended to the at least one user device [104A, 104B, 104C].
At step 414, the method flow [400] ends here.
FIG.5 illustrates a method flow diagram [500] for recommending a sticker, in accordance with an embodiment of the present invention. 25
22
The following includes detailed steps involved for recommending the sticker, wherein the method step initiates at step 502.
At step 504, the sticker is recommended to at least a subset of users. The sticker may also be recommended to all users. In an embodiment, the subset of the user may be the users belonging to a particular geographical location, of a specific gender 5 in a particular age group. Further, the sticker is either recommended randomly or based on one of the first confidence score and the second confidence score. Either of the first confidence score and the second confidence score for the sticker may be kept high for some groups of users and low for other group of users. This strategy, of keeping one of the first confidence score and the second confidence score high for 10 some groups and low for other groups, helps in promoting the sticker to at least a group of users and thus, introducing them among users. In an embodiment, the sticker is a previously stored sticker or a new sticker.
At step 506, the subset of the user is tracked for viewing and selecting the sticker.
At step 508, based on user’s feedback, i.e. whether or not the recommended sticker 15 is viewed and selected by the users of the messaging application, one of the first confidence score, the second confidence score and a combination thereof, is dynamically updated. For an instance if the recommended sticker is the new sticker such as “ola” and the “ola” sticker is selected more number of times than the “Hi” sticker, then the second confidence score associated with new sticker is 20 automatically increased and the first confidence score associated with the “Hi” sticker is automatically decreased accordingly by the training module [210]. The increase and decrease in the value of the confidence scores is based on the user’s feedback. For an instance, if a user selects B sticker from a recommended sticker set of A, B, C and D, then this selection of sticker B, in response to sticker X, is taken as a 25
23
feedback to increase the confidence score of the correlation between the stickers B and X and decrease the confidence score of A, C and D as a response of X. In one embodiment, the quantum of increase or decrease in confidence score may depend on the position of the response sticker in the set of recommendations. For instance, since A is the first sticker in the response set, it is more likely to be selected than D. 5 So, the quantum of decrease in confidence score for A may be greater than that of D. In an embodiment, the response stickers which have high confidence score with respect to a sticker, are recommended to a larger set of users. Further, the response stickers which have a low confidence score with respect to a sticker, are recommended to a smaller set of users or a randomly selected set of users. The 10 present invention also encompasses to update one of the first confidence score, the second confidence score and a combination thereof, based on the tracking of the social-network data of all users. These users are the social-network users present in a social-network system. The social-network system, as described herein, includes the instant messaging application system, the social network platform, and any such 15 system, obvious to a person skilled in the art.
At step 510, the method flow [500] ends here.
FIG.6 illustrates an exemplary screenshot [600] displaying sticker recommendation on the at least one user device [104A, 104B, 104C], in accordance with an embodiment of the present invention. As clearly seen in the screenshot [600], a list 20 of stickers [604] is displayed as recommendation on at least one user device [104A, 104B, 104C]. For example, the recommended list of stickers [604] is provided in response of a “I’m not coming” sticker [602] sent by the first user associated with the least one user device [104A] to the second user. The second user associated with the least one user device [104B] views and/or selects at least one sticker from 25
24
the list of stickers [604] and sends the at least one sticker to the first user associated with the least one user device [104A].
FIG.7 illustrates an exemplary screenshot [700] displaying sticker recommendation on the at least one user device [104A, 104B, 104C], in accordance with an embodiment of the present invention. As clearly seen in the screenshot [700], a list 5 of stickers [704] is displayed as recommendation on the at least one user device [104A, 104B, 104C]. For example, the recommended list of stickers [704] is provided in response of a “Watcha Doin” sticker [702] sent by the first user associated with the least one user device [104A] to the second user. The second user associated with the least one user device [104B] views and/or selects at least one sticker from 10 the list of stickers [704] and sends the at least one sticker to the first user associated with the least one user device [104A].
The present invention also encompasses to recommend the new sticker instead of a response sticker to the at least one user device [104A, 104B, 104C] if said response sticker has become obsolete after a period of time or the user might not like to 15 receive the same response stickers as the recommendation. For example, the first user sends the "Hi" sticker to the second user, the sticker recommendation system [100] recommends a different sticker other than the "Hello", “Hey”, “Hiiiii” sticker/s as the reply stickers and the "Whatsup", "Everything ok", "Are you alive" sticker/s as the follow-up sticker/s. As an example, the “Hello” sticker becomes obsolete after a 20 month. The sticker recommendation system [100] checks one or more new sticker/s that are tagged as "Hi". In particular, a "Moana" movie character sticker is also tagged as a response to the “Hi” sticker. In this case, the second user is provided with "Moana" movie character sticker in reply to the “Hi” sticker. In this particular embodiment, the sticker recommendation system [100] forgets the correlation 25 between the “Hi” sticker and the obsolete “Hello” sticker so that the "“Hello” sticker
25
is not shown as a recommendation for the “Hi” sticker. Moreover, the present invention encompasses to auto-delete any sticker from the sticker suggestion store [208] based on at least one of the updated confidence score and a storage duration of a sticker. For example, if a particular sticker is not used by any user resulting in a low confidence score or the particular sticker is not used by any user for a month, 5 then the sticker suggestion store [208] is configured to auto-delete the particular sticker from the sticker suggestion store [208].
The present invention also encompasses ordering or ranking of stickers within the set of response stickers recommended by the user such that ranking or ordering is based on the first confidence score and/or the second confidence score. 10
The present invention further encompasses to recommend the sticker that are not downloaded by a user. In particular, the sticker which are not stored at the user device [104A, 104B, 104C] or the sticker suggestion store [208] are recommended to the at least one user device [104A, 104B, 104C].
The modules, databases and/or components discussed may be present in the form 15 of a hardware or a software or a hardware-software combination for performing functions and/or operations, as described herein. The connections and/or links between each module, databases and/or component shown in the figures are exemplary and may be connected in other possible ways. The connections and/or links between each module, databases and/or component may be physical (such as 20 wired or wireless connections/links) or logical (such as implementing in semiconductor device).
The present invention provides immense improvement over the existing recommendation system and has numerous advantages. Some of these advantages may include, but are not limited to, efficiently recommending stored and new 25
26
stickers to users based on confidence score. In addition, the present system provides better user experience by recommending new stickers to the users even if the user has not downloaded the new sticker. Moreover, a sticker is auto-deleted from the store based on the usage history of the sticker which in turn keeps the user up-to-date with the stickers which are frequently used or in trend. Furthermore, the 5 invention dynamically updates the confidence score of the sticker based on the tracked usage of the sticker by the user.
According to one embodiment of the present disclosure, the techniques described herein are implemented by one or more special-purpose computing devices. The special-purpose computing devices may be hard-wired to perform the techniques, 10 or may include digital electronic devices such as one or more application-specific integrated circuits (ASICs) or field programmable gate arrays (FPGAs) that are persistently programmed to perform the techniques, or may include one or more general purpose hardware processors programmed to perform the techniques pursuant to program instructions in firmware, memory, other storage, or a 15 combination. Such special-purpose computing devices may also combine custom hard-wired logic, ASICs, or FPGAs with custom programming to accomplish the techniques. The special-purpose computing devices may be desktop computer systems, portable computer systems, handheld devices, networking devices or any other device that incorporates hard-wired and/or program logic to implement the 20 techniques.
The sticker recommendation system [100] may include a bus or other communication mechanism for communicating information, and a processor coupled with the bus for processing information. The hardware processor may be, for example, a general-purpose microprocessor. 25
27
The sticker recommendation system [100] may also include a main memory, such as a random-access memory (RAM) or other dynamic storage device, coupled to the bus for storing information and instructions to be executed by the processor. The main memory also may be used for storing temporary variables or other intermediate information during execution of instructions to be executed by the 5 processor. Such instructions, when stored in non-transitory storage media accessible to the processor, render the computer system into a special-purpose machine that is customized to perform the operations specified in the instructions.
The sticker recommendation system [100] further includes a read only memory (ROM) or other static storage device coupled to the bus for storing static 10 information and instructions for the processor. A storage device, such as a magnetic disk, optical disk, or solid-state drive is provided and coupled to the bus for storing information and instructions.
The sticker recommendation system [100] may be coupled via the bus to a display, such as a cathode ray tube (CRT), for displaying information to a computer user. An 15 input device, including alphanumeric and other keys, is coupled to the bus for communicating information and command selections to the processor. A cursor control, such as a mouse, a trackball, or cursor direction keys, may also be coupled to the bus for communicating direction information and command selections to the processor and for controlling cursor movement on the display. The cursor control 20 typically has two degrees of freedom in two axes, a first axis (e.g., x) and a second axis (e.g., y), that allows the cursor control to specify positions in a plane.
The sticker recommendation system [100] may implement the techniques described herein using customized hard-wired logic, one or more ASICs or FPGAs, firmware and/or program logic which causes the computer system to be a special-purpose 25
28
machine. According to one embodiment, the techniques herein are performed by the computing device in response to the processor executing one or more sequences of one or more instructions contained in the main memory. Such instructions may be read into the main memory from another storage medium, such as the storage device. Execution of the sequences of instructions contained in the 5 main memory cause the processor to perform the process steps described herein. In alternative embodiments, hard-wired circuitry may be used in place of or in combination with software instructions.
The term “store” as used herein refers to any non-transitory media that store data and/or instructions that cause a machine to operate in a specific fashion. Such 10 storage media may comprise non-volatile media and/or volatile media. Non-volatile media includes, for example, optical disks, magnetic disks, or solid-state drives, such as the storage device. Volatile media may include dynamic memory, such as the main memory. Common forms of storage media include, for example, a floppy disk, a flexible disk, hard disk, solid-state drive, magnetic tape, or any other magnetic 15 data storage medium, a CD-ROM, any other optical data storage medium, any physical medium with patterns of holes, a RAM, a PROM, and EPROM, a FLASH-EPROM, NVRAM, or any other memory chip or cartridge.
Various forms of store may be involved in carrying one or more sequences of one or more instructions to the processor for execution. For example, the instructions may 20 initially be carried on a magnetic disk or solid-state drive of a remote computer. The remote computer can load the instructions into its dynamic memory and send the instructions over a telephone line using a modem.
The sticker recommendation system [100] also includes a communication interface coupled to the bus. The communication interface provides a two-way data 25
29
communication coupling to a network link that is connected to a local network. For example, the communication interface may be an integrated services digital network (ISDN) card, cable modem, satellite modem, or a modem to provide a data communication connection to a corresponding type of telephone line. As another example, the communication interface may be a local area network (LAN) card to 5 provide a data communication connection to a compatible LAN. Wireless links may also be implemented. In any such implementation, the communication interface sends and receives electrical, electromagnetic or optical signals that carry digital data streams representing various types of information.
Although the present invention has been described with respect to the concept of 10 recommending stickers, however, it may be obvious to a person skilled in the art, that other forms of visual expressions such as emoticons, GIF images, animations, cartoons, emojis, may also be recommended using the system and the method encompassed by the present invention.
Although, the present invention has been described with respect to a scenario 15 where the first user associated with user device [104A] selects and sends the at least one sticker to the second user associated with user device [104B] and in response to receiving the at least one sticker, the second user selects and sends the at least one response sticker to the first user associated with user device [104A], however, it will be appreciated by those skilled in the art that the present invention is also 20 applicable in scenarios where any end user may select and send any sticker to each other from any user device.
Though a limited number of the at least one user device [104A, 104B, 104C], the first user, the second user, sticker recommendation system [100], the network [102], the stickers, and the link/connection/requests/interfaces/communication, have been 25
30
shown in the figures; however, it will be appreciated by those skilled in the art that the overall system of Figure. 1 of the present invention encompasses any number and varied types of the entities/elements such as the at least one user device [104A, 104B, 104C], the first user, the second user, sticker recommendation system [100], the network [102], the stickers, and the 5 link/connection/requests/interfaces/communication.
While considerable emphasis has been placed herein on the disclosed embodiments, it will be appreciated that many embodiments can be made and that many changes can be made to the embodiments without departing from the principles of the present disclosure. These and other changes in the embodiments of 10 the present disclosure will be apparent to those skilled in the art, whereby it is to be understood that the foregoing descriptive matter to be implemented is illustrative and non-limiting.

We claim:
1. A method, comprising:
- analyzing a social-network data to determine one or more response stickers corresponding to a sticker, wherein said social-network data includes at least a chat history of one or 5 more users;
- based on the analysis, storing a first correlation between the sticker and the one or more response stickers, at a sticker suggestion store [208], wherein said first correlation is automatically assigned a first confidence score, wherein the 10 first confidence score is dynamically updated based on the analysis;
- receiving a new sticker at the sticker suggestion store [208];
- based on the stored first correlation, determining a response sticker set correlated to the received new sticker, wherein a 15 second correlation between the new sticker and the response sticker set is automatically assigned a second confidence score; and
- based on the first confidence score and the second confidence score, recommending, in response to one of the sticker and 20 the new sticker, at least one of said one or more response stickers, said response sticker set and a combination thereof.
2. The method as claimed in claim 1, wherein determining the response sticker set correlated to the received new sticker includes:
32
identifying at least one similar sticker from one or more stickers stored at the sticker suggestion store [208], by comparing a metadata of the new sticker with a metadata of the one or more stickers; and
correlating said one or more response stickers corresponding to the at least one similar sticker, with the new sticker. 5
3. The method as claimed in claim 1, further comprising:
based on one of the first confidence score and the second confidence score, recommending the one of said one or more response stickers, said response sticker set, the new sticker and a combination thereof, to at least a subset of users from a plurality of users registered on a 10 social networking system; and
analyzing whether the subset of users select the recommended sticker at the social networking system.
4. The method as claimed in claim 1 or 3, further comprising:
analyzing the chat history received at the social networking system; 15
based on one of the analysed chat history and the selection of the new sticker, automatically updating at least one of the first confidence score, the second confidence score, and a combination thereof; and
based on the updated confidence score and a storage duration of a 20 sticker at the sticker suggestion store [208], auto-deleting the sticker from the sticker suggestion store [208].
33
5. The method as claimed in claim 1, wherein the recommended sticker is present on at least one of a user device [104A, 104B, 104C], the sticker suggestion store [208] and a combination thereof.
6. The method as claimed in claim 1, wherein the new sticker is at least one of a reply sticker and a follow-up sticker. 5
7. A sticker recommendation system [100] comprising:
- an analysis module [206] configured to
analyze a social-network data to determine one or more response stickers corresponding to a sticker, wherein said social-network data includes at least a chat history of one or 10 more users;
- a sticker suggestion store [208] configured to
store a first correlation between the sticker and the one or more response stickers, based on the analysis, and
receive a new sticker; 15
- a recommendation module [204] configured to
determine a response sticker set correlated to the received new sticker based on the stored first correlation, and
recommend, in response to one of the sticker and the new sticker, at least one of said one or more response stickers, said 20 response sticker set and a combination thereof; and
- a training module [210] configured to
34
automatically assign a first confidence score to said first correlation, wherein the first confidence score is dynamically updated based on the analysis, and
automatically assign a second confidence score to a second correlation, between the new sticker and the response sticker 5 set,
wherein the recommendation is based on the first confidence score and the second confidence score.
8. The system as claimed in claim 7, wherein the recommended sticker is present on at least one of the user device [104A, 104B, 104C], the 10 sticker suggestion store [208] and a combination thereof.
9. The system as claimed in claim 7, wherein the recommendation module [204] is further configured to identify at least one similar sticker from one or more stickers stored at the sticker suggestion store [208], by comparing a metadata of the new sticker with a 15 metadata of the one or more stickers.
10. The system as claimed in claim 9, wherein the analysis module [206] is further configured to correlate said one or more response stickers corresponding to the at least one similar sticker, with the new sticker.
11. The system as claimed in claim 7, wherein the recommendation 20 module [204] is further configured to recommend the one of said one or more response stickers, said response sticker set, the new sticker and a combination thereof, to at least a subset of users from a
35
plurality of users registered on a social networking system, based on one of the first confidence score and the second confidence score.
12. The system as claimed in claim 11, wherein the analysis module [206] is further configured to analyze whether the subset of users select the recommended sticker at the social networking system. 5
13. The system as claimed in claim 7 or 12, wherein the training module [210] is further configured to automatically update at least one of the first confidence score, the second confidence score, and a combination thereof, based on one of the analysed chat history and the selection of the recommended sticker. 10
14. The system as claimed in claim 13, wherein the sticker suggestion store [208] is further configured to auto-delete a sticker from the sticker suggestion store [208], based on the updated confidence score and a storage duration of the sticker.
15. The system as claimed in claim 7, wherein the new sticker is at least 15 one of a reply sticker and a follow-up sticker.