International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 08 Issue: 07 | July 2021 www.irjet.net p-ISSN: 2395-0072
© 2021, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 3565
Mobile Personalized Recommendation of Trends for Social Networks
Gajanan Vinayak Bhole
1
, Tejashwa Khare
2
, Ashutosh Srivastava
3
, Omkar Kadam
4
1
Assistant Professor, Dept. of information technology, Bharati Vidyapeeth University, Pune, Maharashtra, India
2
UG student, Dept. of information technology, Bharati Vidyapeeth University, Pune, Maharashtra, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Number of users and their information feed is in-
creasing day by day on social networks. The users communicate
through social networks, share and distribute information by
sending short text messages in near real-time. As a results of the
social networks, the users are often experiencing information
overload as they interact with many users and stick with
it reading contents on large scale. Recommendation systems are
proposed to assist users accommodate information overload by
prioritizing the things of user’s interest. The user’s preferences
are s haped by personal interests. At the identical time, users
are full of their surroundings, as determined by their geograph-
ically located communities. One among the approach takes into
account both personal interests and native communities. These
community preferences are generally reflected within
the localized trending topics. The proposed dynamic recommen-
dation system provides better customized content described with
the most important tweets on social media like Twitter per
his/her individual interests considering the placement diffusion.
Hence the effect of change within the geographical community
preference on individual user’s interests is observed through this
system which provides top trending tweets supported the identi-
cal
1. INTRODUCTION
Social networks having large growth recently in number
of users and their shared information. The social networks
have different challenges in providing differing types of
information to the user. Twitter, Face-book may be a social
networking applications that allow people to brief about
a broad range of topics. Personalized recommendation often a
promising solution for the knowledge overload problem in
social network sites. Three recommendation problems on
social network sites are explored, being recommending peo-
ple, recommending information, and recommending conversa-
tion Social networks became a vital source for generating
recommendations. Using social networks to understand the
relations between users and their friends as well because
the information obtained about them can improve the
knowledge about users’ behaviours. Also integrating recom-
mendation systems into social networks can provide new
observations and thus decisions that cannot be achieved
through using traditional recommendation systems. Research
studies have also found different properties of social net-
works encourage the combination of recommendation sys-
tems with social network. In this paper, the study is varied and
address areas like network structure, trust, information credi-
bility, event detection, social tagging, Geo fencing etc. The
recommend- systems aim is returning items that are kind of
like the users’ demand. To provide the user with personalized
recommendations for online social network infor-
mation associated with personal and community level using
locations of the user. The system may be show the quantity of
users they will be login within the trend set location. This
technique may be useful for new friend creation. Also during
this system it's easy to identify favorite category for tweets
for the aim of read or write the tweets. The users’ preferences
are shaped by personal interests. At the identical time, users
are plagued by their surroundings, as determined by their
geographically located communities. Ever since the dawn of
civilization, people in general have always been an element of
1 tribe or another, brought together by their shared interests
and a standard thanks to communicate the identical. Capturing
user’s interest, which change over the time, is vital nowadays.
So, specializing in suggestions provided by the social me-
dia must be improved. The social networks suggest the recent
trends to the users supported their location will reflect posi-
tively on their online experience. User can show message the
correspond. Hence, it's important to mine user‘s interest from
social network. Although tweets may contain valuable infor-
mation, many don't seem to be interesting to the users. an
oversized number of tweets can overwhelm users since they
interact with many other users and that they should read ever
increasing content volume on their timeline. Thus, the prob-
lem find the “matching” users and recommending content that
are of interest to users became a key challenge for social net-
works sites. Recommendation systems are proposed to as-
sist users address information overload by predicting the
things that a user is also fascinated by. Recommender sys-
tems became a crucial research area since the looks of the
primary papers on collaborative filtering within the mid-
1990s . There has been much work done both within
the industry and academia on developing new approaches to
recommender systems over the last decade. The inter-
est during this area still remains high because it constitutes a
problem-rich research area and since of the abundance of
practical applications that help user to handle information,
overload and supply personalized recommendations, content,
and services to them. However, despite all of
those advances, the present generation of recommender fur-
ther improvements to form recommendation methods more
practical and applicable to a good broader range of real-life
applications. A way is proposed to spot tweets that will be of
potential interest to the user. Since the user’s interests in
numerous topics change over time, we specialize
in studying this alteration, and recommending to eve-
ry user the foremost interesting tweets on the user’s timeline
at specific time. The users’ preferences are shaped by personal
interests. At the identical time, users are laid low with their
surroundings, as determined by their geographically located