Abstract multimedia & multimedia. The analysis of the user


a days the multimedia social networks plays major role in our daily life.
 All the earlier d MSNs are validated and developed  very well . The past decade has witnessed the
emergence and progress of multimedia social networks (MSNs), which have
explosively and tremendously increased to penetrate every corner of our lives,
leisure and work. As well as, the users are enabled by Mobile internet &
terminals for accessing the MSNs where ever they are and when they want with
the help of any identity. It may be a group or a role. So it become very
complicated & comprehensive to provide the behavior’s interaction between
MSNs as well as in users. The implemented system having the advancements and
developed framework of the analytics in a particular domain; which is called as
SocialSitu, And We implemented an algorithm which is named as novel for
the analysis of the serialized users intention according to the typical GSP
which is the short form of Generalized Sequential Pattern. An enormous
number of users behavior records were broken for exploring the usual sequence
mode. It is mandatory for guessing the intention of the user. We considered the
two types of intentions. Those are playing multimedia & sharing multimedia.
These 2 are widely used in regular MSNs with the help of  intention
serialization algorithm in control of various min support threshold  (Min_Support).
With the help of microscopic behavior analysis of the users, we find out the
each user behavior patterns which are in optimized manner in control of the Min_Support.
Based on the different identities of the user, the behavior patterns of the
users may be varied in session data which is very large.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now


Index Terms—multimedia
social networks, situation analytics, intention prediction, behavior pattern,
big data.


Now a days the we can observe the growth of the users and
contents in multimedia as there are rapid advancements takes place. With the
help of large scale video data set & MSNs, the digital contents are easily
accessed by the users.  As a result, the communication between user &
user and user & system increases.  Hence, providing of personalized
services timely and rapidly may become complex interaction. It become a
challenge in social networks of the multimedia. Usually the computing &
speaking of the multimedia can be classified into three types. First one is
data centric compression of multimedia, communication of the content centric
multimedia & multimedia. The analysis of the user centric social
 media considers the trust modeling of the user and mining paths
propagation & sharing of the digital right ad forensic which is digital in
nature. So that, content users of the multimedia exact needs understanding
& guessing in various situations study is not done well. The first proposal
is CA which is the short form of Context-Aware . Here, the context is defined
in terms of location, nearby people, objects, objects changes all these are
considered a set.  After that the Situ theory proposed by Chang with the
combination of environment of the service using the awareness of the situation
for regulating the changing updates oor run time service developments. Hence,
we reached the user needs which changes and we can offer personalized user
services. The awareness about the situation is necessary for offering the
timely response and good environment in service. then we can easily adapt to
the dynamic service. The human is considered as open and complex system in
social media networks. The intention of one person may change at any time so
that the needs of the user also changed. However, as the behavior of the human
and context is dynamic in nature. The needs of user may changes based on the
dynamic change characteristics. we can say that based on various studies.
  According to the user situation and resources feedback the user
intention can be reflected. According to the intention of the user, the user
personalized services timely can be done by a system. It leads the experience
of the user services. There are various roles in various group for the users in
social media networks. The intention of the user may be changed based on the
various identifications. The users behavior changes is reflected by the
intention changes of the user. Coming to the social media networks, based on
the various identities, user intention analysis may not reach the full accuracy
by Situ theory. The primary motivation of the social media networks sequences
mode user intention analysis. The major advantages of this implemented system
is enrichment as well as advancement of the Situ theory in social domain.
 and second one is proposal of the novel algorithm to analyses the
behavior pattern of the user. The main motto of this implemented system is
prediction of the in depth user intention and earlier work large volume
according to the mental intention. The progress of this work done on the
following sections. The 2nd section having the related works, further section
covers the Situ theory advancement,  next one describes the intention
serialization algorithm & details and its results presented in 4th section.
Final one shows the conclusions. .



situation   analysis
  theory importance and influence as well as Situ   theory
in software engineering studied. The details of the situ system that gives
services of the personal, based on detection of the service update in real
time. Ming et already raised the analysis of the spatial method according to
the theory of Situ. After that the Rahman et al proposed the sharing of the
user data in social circle with friends with the help of a service which is
social. SenseFacen framework used for users service recommendation with
information of multimedia and network of user sensor of the data. Shen et al
stated that the relationship of the social network and environment of the
surroundings. It can able to detect the situation of the user and preferences,
relationship of the users which are nearer  by computing the situation of
the user similarities & guessing the user situation. After  that Zhang
 et  al, advanced N gram prediction algorithm for predicting the web
access request which come in future with log data of the server. Bar-David et
al  proposed guessing the user location based on the earlier objects that
are in moving. If we discard the moving object errors. According to the different
context positions for predicting the vehicle position, context-aware position
prediction algorithm proposed. Lee  et al  designed another algorithm
to allow the smart phones users for accessing the easy services in time. For
predicting the intention of the user and offering the correct services the
condition of the  event  behavior model along with  rule
induction algorithm stated. It is used for finding the mobile users intentions
MSNs. a data of the user noise as well as discrete in MSN. For mining and
analysis this will not be useful. So that we have to gathered the date of the
user preprocess earlier.  The theory of the change situation analytics is
for the software which is not much appropriate for engineering applications.
For providing the best services personally, with context and MSN situations.
For gathering the sequence of the intention of the user I proposed the
algorithm of the novel to analyses the user sequences of SocialSitu(t) with
GSP advancements.


will be huge number of users in various groups in MSNs. They are playing
different role in those groups.  So the desires of the persons may deepen
on the role. The enrichment of the Situ advancement of this framework in social
media, represents in the following section

  1   (Situation(it)):
  At t time the situation can be represented It represents
  the that having the 3ple

, Situation(t)
= {d, A, E}.

the user desire referred by d at t interval,
user action can be represented by A. d is the achievements. At time t,
the context of the environment referred by the. E

(SocialSitu(t)) as the 2nd Definition:  Situation
can be represented by it at

It is the  Situation(t)    extensional
  in the domain which is social in nature. . SocialSitu(t is the 4thple
SocialSitu(t) = {ID ,d,A, E}.

information of the user can be represented by the ID; At t,

desire of user can be represented by d


Definition 3  (ID):
 Identity information of the user can be represented by the ID ; 2ple
ID= {Group, Role}.

should be good relation between the group of the user and role in that group in
MSNs. . .

Definition 4 (Group):
 For a specific reason, sometimes small group are created in social
media.  It is also a part of the network
of the social mediIt’s a part of the whole social media network.

Definition 5 (Role,
R): The role of the user in MSNs.

set R={r1, r2, …, rn},

Definition 6 (Desire,
D): While utilizing the social media , it is the thing what a user need and

{d1, d2,…, dn} ? d i
(1 ? i ? n) are referred to
desire of the user at i.

Definition 7 (Goal,
G): common target of the user’s G={g1,

… , gn}

Definition   8   (Intention,
  I):   It  represents the

user  sequence towards the achieve target 
from the  point  to

initial point I={SocialSitu(1), SocialSitu(2),
 …,  SocialSitu(n)} ? n ? N? SocialSitu(1)

The target can be achieved by the user with
the help of intention sequence in Fig.1.Fig.1. Intention sequence
here, at an particular point SocialSitu(t) represented by every point


. startj ( 1 ? j ? n, j
? N

we may or may not have same are different initial points.

End represents
the closing point Intention(i).. at Intention(i) every sequence, same
node may be terminated except the end point.




The SocialSitu(t) which is frequent at
specific goal. The intention of the user sequence is presented in historical
access record in MSNs.  The database
saves the sequence of the user intention along with a goal which is specific.
By comparing the current user sequence and intention user sequence, we can
guess the present user intention for making the perfect reply for the request
of the user. Then we can provide the service which is personalized. Finding of
the intention of the user is the major issues here.

Fig.2. user flowchart Intention prediction


For finding the relationship among the
different items this rule which is named as association rule in big.
transaction set  can be represented by
DS. here  we have the term
“item” which is nothing but attributes. . data item set, I=
{i1, i2,  …, im},


A ? B

? N
, i ? 1 .

A ? I, B ? I

of  the  association  rule;   A ? B

corresponds to this rule.is  the
 item  set  which

All   frequent   SocialSitu(t)
  related   to   a   certain   goal
achievement in a user’s historical access record consist of

?  Support: item set number  R presented in  DS  named as R
 supporting  number,


R ? S

the number  in the RMSNs item sets. Must
have the minimum single intention sequence. The sequence of the user’s
intention along with a particular specific goal which the database can save.
Comparison of the current user sequence
intention user sequences in database for guessing the present user
intention. R ? S
is t ( R ? S
) = P ( R ? S ).

?Confidence: Rule Confidence R ? S represents
probability where complete data set DS having 
A includes    B     ( R ? S )=P(R|S).

 – tem
set which is frequent. satisfies the  Min_Support


serialization done according to the rule for detecting
the every sequence for respective intent at every ending point

Intention(i) sequence as
taken as  association result rule,. This
helpful for obtaining the association antecedent  rule. The algorithm  of the Intention serialization represents in
Algorithm 1 intention  serialization algorithm flowchart described in
Fig.3. serialization algorithm steps according to rule  of the 
association as below:

(1)  The  database of the web  log
 database can be  scanned  later processing the data   in definition 7 can be  identified association rule  result for generating the Lk+1 sequence  k+1 length,

(4)  3rd Step  repeats till where no longer new
person sequence can be generated. sequences of the  SocialSitu(t) in
relation to Intention(i)  target g’


(5) All sequences of the SocialSitu(t) sequences
respected to database of the target like a user end  point   in
obtained and saved as Intention(i+1). Then, recorded like  G’  = {g’ , g’ ,
   , g’  }, 1 ? m ? n , G ‘
 ? G

the  (2), (3), and (4). steps.

  From step (1) g getting utilized as association result rule.
every SocialSitu(t) utilized association   rule   
antecedent   for computing    support each rule & find
a rule that satisfy Min_Support.

The rule antecedents came with  step (2)

for setting set L1, Lk 
candidate sequence generation is  Ck+1
  k+1 length.

scanned data set, compute with sequence of each candidate like antecedent
& g

till no new goal. Link   operation:  After s1 obtained the
sequence   equal to S2. Where the S2 added to S1.

Pruning  operation:  when
particular candidates  sub-sequence  pattern  is  not equal
to this candidate sequence  pattern. Hence it can be deleted.

1: With the help of situation-aware,
the Intention serialization algorithm

DS  – Dataset

 Min_Support  – Minimum Support:,

‘ G – User’s Goal


Sequence Situ – 

Behavior Analytics

 ( DS,
Min_Support,’ G )

1: Begin

2: for j?1 to n //n represents  user’s goal

 3: for t ?
0 to T

4: Support( ‘ )(j ? gtSocialSitu ) = P (
‘ )(j SocialSitu ?
gt );

         5: endfor

6: if (Support ( ‘ )( ? gtSocialSitu j

7: 1 L ? tSocialSitu ; //the 1-frequent item sets L1

8: endif

9: for k ? 2 to m and k ?1 ?

10: Generate candidate setsCk;

         11: Support ( ‘jk ?
gC ) = P ( ‘jk ?
gC );

12: if (Support ( ‘jk ? gC )>Min_Support)

13: ? CL kk;

14: endif

15: endfor

16: Intention(i) =’jk ? gL ;


17: endfor

18: End

Fig.3. Algorithm of situational aware intention
serialization algorithm Flowchart




CyVOD  is the prototype system in In the multimedia
social network which achieved supports the login of the user, normal quit,
first enters the user goal earlier access of CyVOD, based on  goal range. The system can be quit if the
user achieve goal. We can track sequence of SocialSitu(t).

SocialSitu(t) four elements:

 (1) ID: Role of the user & obtain in
database with  session information that
presented server..

d:  The behavior of the user in MSNs
predictable vector.

User’s behavior A: The behavior of the user may be automatic / compound to get

e: information  of the terminal (mobile
terminal or PC) .

log data is  complex in MSN and it also
irregular. We have to convert th complex to required format. Data cleaning,
processing, identification are involved. represents in in Fig.4,


intention serialization Data preprocessing

Cleaning of Data:
Removed the unnecessary data  from the

identification: Current data of the all users with logs  large quantity identified.

identification  of Session:
pages collection accessed with user in a specific period. Complete session of
the user identified.

transformation of  Data:
Continuous Log data transformers to required data type.

intention of the user is play as well as share data in this. we collect the
play and share  data SocialSitu(t) here.


The awareness of situations become mandatory
in typical MSNs environment.  Environment of the User,
  behavior can be changing as well as the changing of the intention
also may see in individual’s. for adopting the  user changes identities
which are dynamic  in nature and social  domain. Here, we have the Si
which is the extended and   enriched. SocialSitu  theory
is built for  networks of the  social media. serialization
 algorithm  achieved and implemented for social networks which runs
on multimedia. The sequence Moe of the user intention
which is frequent can be obtained with this algorithm. If the identity of the
user changed, we can know the different ID which describes behavior pattern ,
Confidently we obtained the various sequences of the SocislSitu(t), on
same intention if his or her group or role can be changed. Coming to the works
which can be done in future, we can guess the various in depth intentions of the
user buy the help of typical user cloud sequence patterns of the intentions. So
that we can use the SocialSitu. Implemented system is for improving the
multimedia system recommendations as well as MSNs killer applications. .








1     Y. G. Jiang and
J. J. Wang, “Partial Copy Detection in Videos: A Benchmark and an Evaluation of
Popular Methods,” IEEE Trans. Big Data, vol. 2, no. 1, pp. 32-42,
Jan/Mar 2016, doi:10.1109/TBDATA.2016.2530714.

2      C. K. Chang, “Situation
Analytics-A foundation for a new software engineering paradigm,” Computer,
vol. 49, no. 1, pp.

Jan. 2016.


3       P.  Cui,
 W.  Zhu,  T.  S.  Chua,  and  R.
 Jain,  “Social-Sensed

Multimedia Computing,” IEEE Multimedia, vol. 23,
no. 1, pp.

92-96, Jan/Mar 2016


 X. L. Shen, M. K. O. Lee, and C. M. K. Cheung, “Exploring online
behavior in crowdsourcing communities: A relationship management perspective,” Comput.
Hum. Behav., vol. 40, pp. 144-151, Nov. 2014.

       5   C.
Tekin and M. Van Der Schaar, “Contextual online learning for multimedia content
aggregation,” IEEE Trans. Multimedia, vol.     17,
    no.     4,
    pp.      549-561,
    Apr.      2015,

Networks,” Soc. Netw. Anal. Min., vol. 3, no. 4, pp. 969-979, Dec.