HASHTAG ANALYTICS FOR BETTER BUSINESS

INTELLIGENCE

A PROJECT REPORT

Submitted by

TARUN.G (19351064)
HARISH.V(19351074)
JASON PUSHPARAJ(19352002)
RISHI ADHITHYAN(19351055)
in partial fulfillment for the award of the degree

of

BACHELOR OF COMPUTER APPLICATIONS

HINDUSTAN INSTITUTE OF TECHNOLOGY AND SCIENCE

CHENNAI - 603 103

MAY 2022
 BONAFIDE CERTIFICATE

Certified that this project report “HASHTAG ANALYTICS FOR BETTER

BUSINESS INTELLIGENCE” is the bonafide work of TARUN.G (19351064),

HARISH.V (19351074), RISHI ADITHYAN(19351055), JASON PUSHPARAJ

(193520020) who carried out the project work under my supervision during the

academic year 2021-2022.

HEAD OF THE DEPARTMENT SUPERVISOR

DR. P. RANJANA DR. ANGELINE BENITA

Professor and Head Assistant Professor
Department of computing Applications Department of Computer Applications

INTERNAL EXAMINER EXTERNAL EXAMINER

Name: ________________________ Name: _______________________

Designation: _____________________ Designation: __________________

Institution Name: ________________

Project Viva - voce conducted on ___________

I
 ACKNOWLEGEMENT

I wish to express my sincere gratitude to Dr. P. Ranjana, Professor & Head, Department
of Computer Applications, for having evinced keen interest in our project and for her continued
support.
We also thank, Dr. Angeline Benita Asst. Professor (SG), Project Coordinator (BCA),
for her support throughout the completion of our project. We are indebted to our project guide Dr.
Angeline Benita Asst. Professor (SG), Department of Computer Applications for his stimulating
suggestions and encouragement which helped me to coordinate my project especially in writing
this report.

A special thanks goes to all my staff members and friends who helped me in my
coding and gave their valuable suggestions about the project.

Last but not least, I would also like to thank God Almighty and my family who were
keen in helping me in all ways and to give their blessings.

II
 ABSTRACT

As this era, is completely depends upon social media marketing # hashtags are very much
essential part of a contest strategy this little links create method to find out the relevant contents to
share. paving the way for the users to find the proper content for well going of the web sites. As
Business intelligent has the strong procedures, strong technical structures.it enciphers many data
analytics and descriptive analytics.
Requirement hashtag of analytics drives the users to attain the efficiency in
a professional way. With the help of the analytics the search procedures been simplified and
beneficially for the users. In this work, we proposed the model. : Social networks have become a
common part of many people’s daily lives. Users spend more and more time on these platforms
and create an active and passive digital footprint through their interaction with other subjects.
These data have high research potential in many fields, because
understanding people’s communication on social media is essential to understanding their attitudes,
experiences and behaviours. Social media analysis is a relatively new subject. There is still a need
to develop methods and tools for researchers to help solve typical problems associated with this
area.
Social media has become a part of many people’s lives, and the analysis of data that users
create through their interaction with social networks is thus becoming a very current topic for many
areas of research. This article confirms that hashtags, as a specific part of communication, provide
a research opportunity to identify the attitudes and experience of users in social networks in any
given area. Moreover, the SMAHR framework provides a methodological procedure and software
module for working with hashtags A researcher will be able to focus on the subject of research
entirely which uses social network analysis methods to explore social media communication
through a network of hashtags. The results show that social media analysis based on hashtags
provides information applicable to theoretical research and practical strategic marketing and
management applications.

III
 TABLE OF CONTENTS

CHAPTER NO. TITLE PAGE NO.

Abstract IV

List of Tables IV

INTRODUCTION 1

1.1 Overview 1

1 1.2 Different method of social media marketing 1

1.3 digital marketing and internet marketing 2

1.4 Objectivities 5

LITERATURE REVIEW 6

2.1 Review Details 6

2.2 Social Media Marketing Evaluation

6
framework.
2
2.3 The Influence of Interactive Social Media
6
Marketing.

2.4 Location-Based engagements in Media 7

2.5 Tweet Classification Using Hashtags 7

SYSTEM DESIGN 9

3.1 Existing system 9

3.2 Proposed system 10

3
3.3 Software requirements 11

3.4 Hardware requirements 12

3.5 Diagram 13

IV
 MODULE DESCRIPTION 18

4.1 Data collection 18

4 4.2 Data processing 18

4.3 Hashtag analysis 19

4.4 Modification of the account 19

IMPLEMENTATION 21
5
5.1 Hashtag analyzing code 21

CONCLUSION& FUTURE WORK 31

6 6.1 Conclusion 31

6.2 Future work 31

REFERENCES 36

V
 CHAPTER-1

INTRODUCTION

1.1 OVERVIEW:
Social media marketing sites have taken over today’s world. From what was started merely for
entertainment. Today these websites provide a host of opportunities to its Users ranging from
marketing their products increasing awareness on social issue, sharing photos, video class, joining
groups and forms of interests promoting an idea, finding family roots finding a lost friend, gaming,
entertainment application and so on. In fact, the usage of social media marketing sites has been so
diverse that many people use it for more than just entertainment. Mental (2012) studied the
readiness of health students to use Facebook as a platform for training professional habits
particularity in the case of influenza vaccination. This clearly shows social networking sites serve
as a creative solution to many business problems.

Now that you have an idea of the prevalence of social media and
its rapidly growing influence, let’s take a look at some of the key ways in which social media can
help you connect with your target audience. Generate awareness of your company and its products
or services by tapping into one of the largest audiences in the world. Generate leads through your
social networking connections, speeding up the process by creating events and promotions.

Draw in visitors to your websites and other online portals by using your social media
resources to promote new content. Maintain direct relationships with your customers and potential
customers in order to improve your products and get to better know your audience. Provide
customer service and advice directly to your audience and get useful feedback from your followers

1
Local businesses can also make extensive use of social media platform targeted specially towards
traditional high-street businesses. Sites like Google Places, Bing Places, Yelp and Foursquare,
among others, help you to show up in local search results and online or mobile maps. What’s more
is that all of the above is absolutely free to use, and while there are also paid SMM methods, the
free ones are far too valuable to neglect. Historically, online marketing was a fairly one sides
approach. Business pushed ideas out and consumers passively received them. But that landscape
has shifted. The internet has become extremely interactive. Social media has created a new style
of communication and there are now billions of conversations happening online. People are
discussing popular news articles, sharing photos of their pets, and even engaging with brands. And
it all feels relatively natural to the consumer. It’s just part of how the web operates. All of these
conversations present exciting opportunities for marketers. We can join in on a conversation to
drive brand awareness, or create our own conversations and empower our customers to do the
marketing for us.

Social media marketing is all about creating that goal is to gain traffic to our
websites or attention on our brand. But unlike other interactive moments with our customers to
achieve a goal we’ve defined. Most often forms of marketing, social media taps into the idea of
using your customers as a marketing vessel. The content that you share will hopefully be liked,
retweeted, shared again, blogged about, and so on. This viral effect is what makes social media
marketing so effective. A simple like on a post could expose that content to hundreds of potential
customers you normally wouldn’t have access to. Because social media is so personal, when
friends share content, it comes with another layer of credibility. Done right, social media has the
potential to transform your business. However, it’ll require a good strategy, some creativity, and a
little bit of luck. Done wrong and social media might bring unwanted attention which could
potentially harm your business or your brand image.

We’re going to focus mainly on the organic aspects of social media marketing.
More specifically, the unpaid and natural approach to distributing content. For many brands, social
media will be a must have component of your digital marketing strategy. Your social media might
feature the major networks, or it could be as simple as blog, a customer forum or a niche
bookmarking site will be focusing on marketing with the big two: Facebook and Twitter. Now,
these two networks are not the same. Each is unique and has its own best practices, style and
audience, and we’ll explore their differences together in a later movie. Social media is a moving
target. If you’re able to adapt and scale alongside of it, your brand will benefit in the long term.

1.2 DIFFERENT METHODS OF SOCIAL MEDIA MARKETING

Social network marketing is popularly called as Internet marketing. Today you can find many
ways for internet marketing. Many people who enter this online marketing are less worried
because of its guaranteed success. If you see in Internet totally all types of products have been
marketed online without much effort. Internet attracts many business people to promote their
business online. Social network marketing is grown to such a height that today many people can’t
earn without it. Some of the most recognized network marketing tools are Face book, Myspace
and Linked In. Twitter became regular place for people who have newly entered the field of social
networking marketing.

1. BLOGGING:

When you start Blogging or posting your data about any product, you can see less response from
clients. Later it will become big business via blog. Websites and blogs are most powerful tools
for social network marketing when matched with other networking tools. Blog is an amazing tool
which provides a many other facilities in addition to just marketing your business. It also helps
you to communicate with other clients in case if you have any problems.

2.WEB DESIGN:

It is important to have private website if you are a freelancer. Your website will help your clients
to know about you and it will make them clear that you are a serious freelance marketer and help
to make huge revenue via online marketing.

3.ARTICLE SELLING: It is also best and cheap internet marketing method. It is a mode of
advertising our trade just by writing article and attracting endless number of users across world.
We usually sell our article to different article database websites and articles directories. Today it

2
provided free business to many advertisers and publishers and they are really benefited through
their articles.

4. EMAIL SENDING:
Electronic mail sending is the best way to marketing. Collect list of email address through portfolio
websites and email about your business to all internet users. Your Email should be attractive in
such a way that your recipient will be impressed to get back to you.

5. USE SOCIAL NETWORKING WEBSITES:

Social networking websites like Twitter, face book can be used to promote your sales. These
provide best platform for all who are thinking of online marketing.

6. VIDEO PROMOTION:
Use several video distribution websites for your marketing. These websites upload your service to
the whole world. All that you need to do is film a video about marketing and send it to video
uploading sites like You Tube. It seems it is the easiest way of marketing than any other modes
since many people will be interested in view videos rather than word form of advertisement.

1.3DIGITAL MARKETING AND INTERNET MARKETING

The fields of marketing and advertising have changed over the years and it is clear that that have
always taken advantage of the existing technologies. In the past, when the concept of the internet
was still unimaginable to many, most business capitalized on traditional channels of marketing
such as television, radio, print, and billboard ads. Yes, they are still used by many companies
today and if used effectively, they can give big income returns to your business.

When the internet

was born it transformed the world into a global village. People in different time zones can now
communicate in a click or press of a button. The internet has become so diverse that even things
from your imagination can be found in some corners of World Wide Web. Today, there are two
billion internet users thus, it gave way to the conception of new breeds of marketers and
advertisers.

3
 Digital marketing refers to the use of electronic device such as computers, tablets,
smart phones, and digital billboards, among others to interact with consumers. It can also be
defined as the promotion of brands, products services using different forms of digital media to
reach and engage customers in a timely, relevant personal and cost-effectively way. Compared to
traditional marketing digital marketing is less expensive and you can have instant access to
tangible feedback and reports about saw or heard the ad, and took action. With digital marketing,
you can how a campaign performs in real-time what is being viewed, how often, duration, and
even statistics such as sales conversions.

The internet plays a significant role in digital marketing

which is why people tend to assume that it is synonymous with Internet Marketing. Nevertheless,
they are somewhat different in the sense that internet marketing is just one of the categories of
digital marketing. Internet marketing involves marketing approaches such as search engine
optimization (SO), social media marketing, and pay-per click, to mention a few, while the use of
podcasts, fax broadcasts, text blasts, instant messaging, mobile apps and digital telephony and
video, among others, as marketing channels narrow down the definition of internet marketing.

INDIA DIGITAL MARKETING STATISTICS –2016

As per a recent report by Boston consulting group India has one of the largest fastest growing
populations of Internet users in the world –190 million as of June 2014 and growing rapidly. India
already has the third largest Internet population in the world today, after China with 620 million
and the US with 275 million.

addition, India already has 100 million active Facebook users today,
the second largest number after the US and Canada with 152 million. As one fast forwards, it is
estimated that there will be over 500 million Internet users in India by 2018 –making India the
second largest population of Internet users in the world. The growth in the Internet base in India
has been exponential. It took 20 years from the introduction of the Internet to reach 100 million
users. The second 100 million will likely be reached within three years, and the third in less than
a year. In fact, the next four years will see nearly 350 million additional Internet users. In The
primary drivers of this exponential growth have been expansion to small town / rural
India, increased affordability due to lower cost data enabled handsets, and the ever-increasing
awareness of, and need for, the Intern.

4
 1.4 OBJECTIVES
It is truly said that “A study without objectives is like a tree without roots”. In
any area of research, the first and the foremost task is to decide and define the objectives of the
research i.e., the reason why the research study should be conducted. A research study may have
many objectives, but all these objectives revolve around one major objective, which is the focus
of the study. In this study, the prime objective is to identify opportunities and challenges of the
marketing through the social networking sites. This study is based on emergence of social
networking sites as an efficient marketing tool.

In order to pursue the afore mentioned prime objectives, some allied

objectives have also been identified which are as follows: To figure out the profile of users of
social networking sites. To study the trends and shifts in marketing practices from traditional
marketing to online marketing. To find out the opportunities available to marketers in marketing
through social networking sites. To study the challenges faced by the marketers adopting social
networking sites as their marketing tool. To study the scope of social networking sites to be used
as a marketing.

5
 CHAPTER-2

REVIEW OF LITERATURE:

2.1 Review Details:

Review of literature is a written overview of major writings and other sources on a selected
topic. This provides a critical review of miscellaneous studies, researches, books, scholarly
articles, blogs and all other sources related with social media marketing strategies

2.2 Social Media Marketing Evaluation framework.

AUTHOR: Brendan James Keegan and Jennifer Rowley (2017)

Contributes to knowledge regarding social media marketing strategy by developing a stage
model of SMM evaluation and uncovering the challenges in this process. The research paper has
developed a Social Media Marketing Evaluation framework.
This framework has the following six stages: setting evaluation objectives, identifying key
performance indicators (KPIs), identifying metrics, data collection and analysis, report generation
and management decision making.
Moreover, the paper also identifies and discusses challenges associated with each stage of the
framework with a view to better understanding decision making associated with social media
strategies. Two key challenges depicted by the study are the agency-client relationship and the
available social analytics tools.

2.3 The Influence of Interactive Social Media Marketing.

AUTHOR: Rodney Graeme Duffett (2017)
Examines the influence of interactive social media marketing communications on teenagers‟
cognitive, affective and behavioral attitude components in South Africa. The paper also studies the
impact of a number of additional factors such as usage (access, length of usage, log-on frequency,
log-on duration and profile update incidence) and demographic (gender, age and population group)
variables on young consumers‟ attitudes toward social media marketing communications.

6
The study ascertained that social media marketing communications had a positive influence on
each attitude component among adolescents, but on a declining scale, which correlates to the
purchase funnel model. Thus this investigation also makes an important contribution to attitudinal
research in developing countries, where there is a lack of research in social media marketing
communications.
The practical implication of the study is that the companies and their brands should consider using
and/or adapting their strategies based on the declining impact of social media marketing
communications on the hierarchical attitude stages among young consumers and the divergent
influence on usage and demographic variables when targeting the lucrative and technologically
advanced, but capricious, Generation Z consumers.

2.4 Location-Based engagements in Media

AUTHOR: Priyanka P.V and Padma Srinivasan (2015)
In her research study identified various factors that determine the purchase of a product using
social media from a customer's point of view. A model from the retailer's perspective has been
developed that explains how social media can be used for increasing customer loyalty. The study
concludes that continuous customer support services will result in improvement of customer
retention. This information plays very important role for developing your personal account and
also to increase the followers and likes. Usually, a beginner always wastes his money in order to
promote his/her Instagram account. Moreover, they also use ‘n’ number of hashtags while
uploading a post on their Instagram account just to collect more views and likes.
New applications and social platforms will flourish and allow even greater personalization and
real-time, location-based engagements in media.

2.5 Tweet Classification Using Hashtags

AUTHOR: Vibhuti Gupta and Rattikorn Hewett(2020)
Along with scalable and automated techniques for tweet topic classification using hashtags, there
is also a requirement for real-time analytics approaches to handle huge and dynamic flows of
textual streams generated by Twitter. To address these problems, this paper first presents a novel
semi-automated technique that derives semantically relevant hashtags using a domain-specific

7
knowledge base of topic concepts and combines them with the existing tweet-based-hashtags to
produce Hybrid Hashtags. Further, to deal with the speed and volume of Big Data streams of
tweets, we present an online approach that updates the preprocessing and learning model
incrementally in a real-time streaming environment using the distributed framework, Apache
Storm. Finally, to fully exploit the batch and stream environment performance advantages.

8
 CHAPTER-3

SYSTEM DESIGN

3.1 EXISTING SYSTEM:

In this era, Instagram has various business accounts but these applications are not giving
enough information for the beginners. The information comprises of username, Bios, hashtags,
description, reels, etc.
This information plays very important role for developing your personal account and also
to increase the followers and likes. Usually, a beginner always wastes his money in order to
promote his/her Instagram account. Moreover, they also use ‘n’ number of hashtags while
uploading a post on their Instagram account just to collect more views and likes.
Though Instagram has the facility of providing insights, it is not much useful for the users’
social fame. They can be collected after the analyse of the hashtags then the relevant data for
modelling, removes noisy, low-quality data.
We can perform advanced analytics: sentiment analyses, topic modelling, social network
analyses, trend analyses . If we search according to their perspective results can be taken, hashtags
can be played a major role for analysing in social media to gather attain information about them.
After collecting the data, we analyse the results and create the own professional and trendy
hashtag, description, bio for a type of accounts.
The method of data transformation depends on the software used for data mining. After
collecting data, we are going to process or remove non hashtag text, it is necessary to perform data
filtering based on the selected region.
For this step, it is necessary to transform the data into a form where it will be possible to
identify the selected region. The data can be imported into datasets for others to use, based on the
earth relationship in given message and hashtags can be viewed.
To resolve those issues, we can implement the following methods to increase sales, likes,
followers.

9
3.2 PROPOSED SYSTEM
BIO:
An Instagram bio is the small area under your username where you can share some
details about yourself or your brand. Your Instagram bio can include a brief self or brand
description, contact information, emojis, hashtags and more.
USERNAME:
A social media handle is a public username used on social media accounts. On platforms
like Twitter and Instagram, it's the username followed by an @ symbol, such as @guardian, or
@BBC. But on Facebook, it's the username following facebook.com in the address bar when
someone opens your profile.
With these two others and our friends can find and socialise each other to make that
accounts more professional and famous way using our algorithms.
DESCRIPTION:
The term social media refers to a computer-based technology that facilitates the
sharing of ideas, thoughts, and information through virtual networks and communities. Social
media is internet-based and gives users quick electronic communication of content, such as
personal information, documents, videos, and photos.
Now that you have an idea of the prevalence
of social media and its rapidly growing influence, let’s take a look at some of the key ways in
which social media can help you connect with your target audience. Generate awareness of your
company and its products or services by tapping into one of the largest audiences in the world.
Generate leads through your social networking connections, speeding up the process by creating
events and promotions.
Social media marketing is all about creating
that goal is to gain traffic to our websites or attention on our brand. But unlike other interactive
moments with our customers to achieve a goal we’ve defined. Most often forms of marketing,
social media taps into the idea of using your customers as a marketing vessel. The content that you
share will hopefully be liked, retweeted, shared again, blogged about, and so on. This viral effect
is what makes social media marketing so effective. A simple like on a post could expose that

10
content to hundreds of potential customers you normally wouldn’t have access to. Because social
media is so personal, when friends share content, it comes with another layer of credibility. Done
right, social media has the potential to transform your business. However, it’ll require a good
strategy, some creativity, and a little bit of luck. Done wrong and social media might bring
unwanted attention which could potentially harm your business or your brand image.

Social network marketing is popularly called as Internet marketing. Today you

can find many ways for internet marketing. Many people who enter this online marketing are less
worried because of its guaranteed success. If you see in Internet totally all types of products have
been marketed online without much effort. Internet attracts many business people to promote their
business online. Social network marketing is grown to such a height that today many people can’t
earn without it. Some of the most recognized network marketing tools are Face book, Myspace
and Linked In. Twitter became regular place for people who have newly entered the field of social
networking marketing.

3.3 SOFTWARE REQUIREMENTS:

The functional requirements or the overall description documents include the

product perspective and features, operating systems and operating environments,
graphics requirements, design constraints and user documentation.
The appropriation of requirements and implementation constraints gives the
general overview of project.
- Anaconda
- Python idel 3.7 version
- Java
- SQL
- Android studio.

11
3.4 Hardware requirements:
Pc specifications:
- OS (windows10)
- Intel core i3
- 5th gen
- 4 GB RAM.

12
3.5 DIAGRAMS:
ARCHITECTURE DIAGRAM:

13
UML DIAGRAM:

The Unified Modelling Language (UML) is a diagramming notation (language) that lets you
visualize systems and software. It’s one of the most popular forms of diagramming in software
development and became an ISO standard in 1997. Since then, there have been a few updates to
add extra UML diagram types and to support new technologies in programming.

Diagrams are easier and faster to understand than text, so they are more appropriate to document
systems than comments in millions of lines of code, although these are also important! They help
you and your companies in many situations, like when you …

• onboard new colleagues or team members

• debug to find problems
• plan for new developments or processes
• optimize existing systems and programs
• audit your processes and apply for accreditations

And of course, when you want to develop something new – you will find many problems
beforehand in your UML diagrams that could become showstoppers during development or
quality testing.
There are two main categories and 14 different types of UML diagrams, each of which is used in
a different situation.

• Structure diagrams: show the static relationships between the components in the system.
• Behaviour diagrams: show how the components in the system react to each other, they
capture how the system changes, and in some diagrams, how it changes over time.

14
USE CASE DIAGRAM:

15
ACTIVITY DIAGRAM:

16
SEQUENCE DIAGRAM:

17
 CHAPTER-4

MODULES DESCRIPTION

4.1 MODULE 1: Data Collection

A process which in this project we have created an algorithm which collects information
or data about hashtags, bio, description, reels, etc... we started with a small number of seed
hashtags and use them to collect and label tweets with five prominent emotion classes. The hashtag
extracted from this unlabeled tweet were scored and ranked to extracted high ranked hashtags,
which was then used to form an emotion hashtag repository.
A key limitation of these approaches is that
they can be easily applied for identifying sentiments since their semantics can be easily captured
by a single hashtag (e.g., #sad, #happy) but they are not suitable for identifying general topics (e.g.,
#health, #entertainment) whose semantics require diverse set of hashtags .Some research uses
ontology to classify long documents and for Twitter sentiment analysis. In ontology is used to
identify appropriate classes for multilabel classification of economics documents in various
categories. Work in uses ontology for more elaborate sentiment analysis of Twitter posts, instead
of determining sentiment polarity of tweets.
None of these approaches uses ontology to identify appropriate
hashtags for tweet topic classification. Due to the massive volume of data generated by Twitter, a
significant amount of research has been undertaken in designing large-scale systems for sentiment
analysis. Research in used all hashtags and emoticons as sentiment labels and exploited various
pattern, words, and punctuation features to classify tweet sentiment are collected using python
language which we are created.

4.2 MODULE 2: DATA PROCESSING

Transforming data into a suitable form, which is required by the information

mining system. Basically, it is about preparing data for an information system working with data.
The method of data transformation depends on the software used for data mining. After collecting
data, we are going to process or remove non hashtag text, it is necessary to perform data filtering

18
based on the selected region. For this step, it is necessary to transform the data into a form where
it will be possible to identify the selected region. The data can be imported into datasets for others
to use, based on the earth relationship in given message and hashtags can be viewed. Thus, the
algorithm can be performed to extract and taken for many purposes. And not only for hashtags,
description, bio, reels can be analysed.

4.3 MODULE 3: ANALYSING HASHTAGS

In social media has not implemented any hashtag analysing systems which is suggests
or can’t get any tools to analyse. By using this algorithm, we can get about tracking hashtags,
description, bio information for customers to use. Unlike item -to- item collaborative filtering, the
content-based approach makes use of item content only to determine similarities between them.
They can be collected after the analyse of the hashtags then the relevant data for
modelling, removes noisy, low-quality data. We can perform advanced analytics: sentiment
analyses, topic modelling, social network analyses, trend analyses etc... If we search according to
their perspective results can be taken, hashtags can be played a major role for analysing in social
media to gather attain information about them.
After collecting the data, we analyse the results and create the own professional
and trendy hashtag, description, bio for a type of accounts. The analysing can take little amount of
time which to show the results like (10-15seconds). Comparing to other algorithms like decision
tree, clustering, best peer. In our algorithm have advantage of showing accurate results

4.4 MODULE 4: MODIFICATIONS OF ACCOUNTS

For the beginners in social media, they won’t give any suggestions, algorithms, tools, etc...
in clustering genetic algorithm, hashtags can be clustered and chromosome representation had been
happened. So, the time taken and accuracy are not good to attain more information. Similarly best
peer algorithm has disadvantage of only hashtag and it can get too many errors. Comparing to
them in algorithm wise or in extracting information we have great advantage of time taken is
reduced, accuracy is more perfect, not only hashtags: description, bio, reels can be analysed and

19
suggestions will be given. On other algorithms to understand the UML diagrams and architecture
diagrams are not good to see, in this project we have clearly mentioned both diagrams to
understand. So, it won’t be a problem to search away anywhere. And not only in twitter, in social
medias hashtags, description, bios can be analysed using our algorithm.

20
 CHAPTER-5

IMPLEMENTATION

INTRODUCTION :
When it comes to building large, performant and reliable web applications, Python is a
solid choice. It's trusted by developers at the largest of tech companies like Google, Facebook,
Instagram, Twitter, and Uber.
Python is a high-level, general-purpose programming language. Its design philosophy
emphasizes code readability with the use of significant indentation. Its language constructs and
object-oriented approach aim to help programmers write clear, logical code for small- and large-
scale projects.

5.1 HASHTAG ANALYSING CODE :

import tweepy
from tweepy import OAuthHandler
from textblob import TextBlob
import matplotlib.pyplot as plt
class TwitterClient(object):

def _init_(self):

consumer_key = 'jbkZ2RmduoH7eh9WBpPUQWro3'
consumer_secret = 'RpdmaKXShnzz8NdjlBG6vad88E3CebNtx5Gb03fkEvGCst1fIo'
access_token = '585362387-kqPJV0ztt0Q1RVQdz77LTMhWV0CzNgk1cAJwmBHm'
access_token_secret = 'jYfkksmxZybdn9pwpArAVWzSDpe0J4yaafU3EqQS9yJQ3'

try:

21
self.auth = OAuthHandler(consumer_key, consumer_secret)

self.auth.set_access_token(access_token, access_token_secret)

self.api = tweepy.API(self.auth)
except:
print("Error: Authentication Failed")
def clean_tweet(self, tweet):

return ' '.join(re.sub("(@[A-Za-z0-9]+)|([^0-9A-Za-z \t])|(\w+:\/\/\S+)", " ", tweet).split())

def get_tweet_sentiment(self, tweet):
class _LogInPageWidgetState extends State<LogInPageWidget> {
TextEditingController textController1;
TextEditingController textController2;
bool passwordVisibility1;
TextEditingController textController3;
TextEditingController textController4;
bool passwordVisibility2;
TextEditingController textController5;
bool passwordVisibility3;
final scaffoldKey = GlobalKey<ScaffoldState>();

@override
void initState() {
super.initState();
textController1 = TextEditingController();
textController2 = TextEditingController();
passwordVisibility1 = false;
textController3 = TextEditingController();

22
 textController4 = TextEditingController();
passwordVisibility2 = false;
textController5 = TextEditingController();
passwordVisibility3 = false;
}

@override
Widget build(BuildContext context) {
return Scaffold(
key: scaffoldKey,
backgroundColor: FlutterFlowTheme.of(context).primaryBackground,
body: GestureDetector(
onTap: () => FocusScope.of(context).unfocus(),
child: Stack(
children: [
Image.asset(
'assets/images/space-1.jpg',
width: MediaQuery.of(context).size.width,
height: MediaQuery.of(context).size.height * 1,
fit: BoxFit.cover,
),
Padding(
padding: EdgeInsetsDirectional.fromSTEB(15, 300, 15, 0),
child: DefaultTabController(
length: 2,
initialIndex: 0,
child: Column(
children: [
TabBar(
labelColor:
FlutterFlowTheme.of(context).primaryBackground,

23
 labelStyle: FlutterFlowTheme.of(context).bodyText1,
indicatorColor: Color(0xFFEA0C0C),
tabs: [
Tab(
text: 'SIGN IN',
),
Tab(
text: 'SIGN UP',
),
],
),
Expanded(
child: TabBarView(
children: [
Column(
mainAxisSize: MainAxisSize.max,
children: [
Padding(
padding:
EdgeInsetsDirectional.fromSTEB(0, 15, 0, 0),
child: TextFormField(
controller: textController1,
obscureText: false,
decoration: InputDecoration(
labelText: 'Email Address',
hintText: 'Enter your Email Address',
enabledBorder: OutlineInputBorder(
borderSide: BorderSide(
color: Color(0x00000000),
width: 1,
),

24
 borderRadius: BorderRadius.circular(10),
),
focusedBorder: OutlineInputBorder(
borderSide: BorderSide(
color: Color(0x00000000),
width: 1,
),
borderRadius: BorderRadius.circular(10),
),
filled: true,
fillColor: FlutterFlowTheme.of(context)
.primaryBackground,
),
style: FlutterFlowTheme.of(context).bodyText1,
),
),
Padding(
padding:
EdgeInsetsDirectional.fromSTEB(0, 15, 0, 0),
child: TextFormField(
controller: textController2,
obscureText: !passwordVisibility1,
decoration: InputDecoration(
labelText: 'Password',
hintText: 'Enter your Password',
enabledBorder: OutlineInputBorder(
borderSide: BorderSide(
color: Color(0x00000000),
width: 1,
),
elif analysis.sentiment.polarity == 0:

25
return 'neutral'
else:
return 'negative'
def get_tweets(self, query, count = 20000):

tweets = []
try:

fetched_tweets = self.api.search(q = query, count = count)

for tweet in fetched_tweets:

parsed_tweet = {}

parsed_tweet['text'] = tweet.text

parsed_tweet['sentiment'] = self.get_tweet_sentiment(tweet.text)

if tweet.retweet_count > 0:

if parsed_tweet not in tweets:

tweets.append(parsed_tweet)
else:
tweets.append(parsed_tweet)

return tweets
except tweepy.TweepError as e:

print("Error : " + str(e))

def plot():

26
api = TwitterClient()

tweets = api.get_tweets(query = raw_input("enter the person or word you want to analyze\n"),

count = 20000)
analysis = TextBlob(self.clean_tweet(tweet))

if analysis.sentiment.polarity > 0:
return 'positive'
elif analysis.sentiment.polarity == 0:
return 'neutral'
else:
return 'negative'
def get_tweets(self, query, count = 20000):

tweets = []
try:

fetched_tweets = self.api.search(q = query, count = count)

for tweet in fetched_tweets:

parsed_tweet = {}

parsed_tweet['text'] = tweet.text

parsed_tweet['sentiment'] = self.get_tweet_sentiment(tweet.text)

if tweet.retweet_count > 0:

if parsed_tweet not in tweets:

tweets.append(parsed_tweet)

27
else:
tweets.append(parsed_tweet)

return tweets
except tweepy.TweepError as e:

print("Error : " + str(e))

def plot():
api = TwitterClient()

tweets = api.get_tweets(query = raw_input("enter the person or word you want to analyze\n"),

count = 20000)
ptweets = [tweet for tweet in tweets if tweet['sentiment'] == 'positive']
pt=format(100*len(ptweets)/len(tweets))
print("POSITIVE TWEET PERSENTAGE: ",pt)
ntweets = [tweet for tweet in tweets if tweet['sentiment'] == 'negative']
nt=format(100*len(ntweets)/len(tweets))
print("NEGATIVE TWEET PERSENTAGE: ",nt)
nut=format(100*(len(tweets) - len(ntweets) - len(ptweets))/len(tweets))
print("NEUTRAL TWEET PERSENTAGE",nut)
print("\n\nPOSITIVE TWEETS:")
for tweet in ptweets[:10]:
print(tweet['text'])
print("\n\nNEGATIVE TWEETS:")
for tweet in ntweets[:10]:
print(tweet['text'])
choice=raw_input("PIECHART\n")
if choice=="" :
labels = 'Positive', 'Negative', 'Neutral'
sizes = [pt,nt,nut]

28
colors = ['green', 'red', 'blue']
explode = (0.1,0.1,0.1) # explode 1st slice
plt.pie(sizes, explode=explode, labels=labels, colors=colors,
autopct='%1.1f%%', shadow=True, startangle=40)
plt.axis('equal')
plt.show()
def main():

count = 20000)
ptweets = [tweet for tweet in tweets if tweet['sentiment'] == 'positive']
pt=format(100*len(ptweets)/len(tweets))
print("POSITIVE TWEET PERSENTAGE: ",pt)
ntweets = [tweet for tweet in tweets if tweet['sentiment'] == 'negative']
nt=format(100*len(ntweets)/len(tweets))
print("NEGATIVE TWEET PERSENTAGE: ",nt)
nut=format(100*(len(tweets) - len(ntweets) - len(ptweets))/len(tweets))
print("NEUTRAL TWEET PERSENTAGE",nut)
print("\n\nPOSITIVE TWEETS:")
for tweet in ptweets[:10]:
print(tweet['text'])
print("\n\nNEGATIVE TWEETS:")
for tweet in ntweets[:10]:
print(tweet['text'])
choice=raw_input("PIECHART\n")
if choice=="" :
labels = 'Positive', 'Negative', 'Neutral'
sizes = [pt,nt,nut]
colors = ['green', 'red', 'blue']
explode = (0.1,0.1,0.1) # explode 1st slice
plot()

29
if _name_ == "_main_":

main()

30
 CHAPTER-6

CONCLUSION& FUTURE WORK

6.1 CONCLUSION:

Tweet topic classification using hashtags is a challenging task due to frequent evolution and non-
standardized expressions (i.e., ambiguous, slang, abbreviations) of hashtags by users. Existing
approaches, using external lexical resources for deriving semantics from hashtags, suffer from the
issues of limited coverage and lack of scalability and generality. In this paper, a novel Hybrid
Hashtags approach is presented. The approach derives semantically relevant hashtags using a
domain-specific knowledge base of topic concepts. We first evaluate the effectiveness of the
proposed HHTC-Batch environment where we compare results of our approach with other state-
of-the-art tweet topic classification techniques. The results show that our proposed technique is
effective for tweet topic classification and can be used in combination with some existing
techniques to build a classification model, even from a limited labeled training data set.
Specifically, we propose a technique that combines tweet hashtags with enhanced ontology-driven
hashtags and show by experiments that the technique is effective. Further, to deal with large data
streams, an online technique (HHTC-Stream) is presented and implemented in a real-time
distributed framework, Apache Storm. Finally, a comprehensive framework, Hybrid Hashtag-
based Tweet topic classification (HHTC) is presented where we effectively combine both batch
and online mechanisms. Experimental results show that the combination of batch and online
mechanisms helps to improve classification performance but takes longer to execute. However,
the online approach drastically reduces the execution time, which is helpful for real-time analytics
applications.

6.2 FUTURE WORK:

Overall, our application is beneficial for faster application requirements
while HHTC-Batch and HHTC can be used for more accurate model-building requirements. Both
HHTC-Stream and HHTC improve throughput, speed, and execution time in a multiprocessor
environment. Although we studied a binary tweet topic classification problem in this work, the

31
proposed approach is also general enough to be applicable to non-binary classification. In theory,
like many machine learning techniques, having a large number of classes should not impact the
accuracy, if the training data consists of a well-balanced class distribution that is large enough to
derive valid results. However, in practice, this may not be easy to obtain due to the rarity of the
use of some hashtags. Thus, to deal with such a situation in practice requires a large set of labelled
data. In our future work, we will apply our proposed technique to multi-class classification
problems and further evaluate it with more diverse datasets. Finally, It can be said that using our
application it will lead future generations to gain anything using accounts they will become
successful.

32
33
34
35
 REFERENCES

1. Breivik, E., & Thorbjørnsen, H. (2008). Consumer Brand Relationships: An Investigation

Of Two Alternative Models. Journal of the Academy of Marketing Science, 36(4), 443–
472. doi:10.1007/ s11747-008-0115-z
2. Brogan, C. (2010). Social Media 101. Hoboken, NJ: Wiley. doi:10.1002/9781118256138
3. Bryan, D., & Bryan, D. (2014). Social Media PPC: Facebook Ads vs. LinkedIn Ads vs.
Google AdWords, which would you choose? Opace.co.uk. Retrieved June 20, 2014, from
http://www.opace. co.uk/blog/social-media-ppc-facebook-ads-vslinkedin-ads-vs-google-
adwords-which-wouldyou-choose
4. Campbell, A. (2010, Jan 21). Amy Campbell’s Web Log » Social Media — A Definition.
Retrieved June 16, 2014, from https://blogs.law.harvard. edu/amy/2010/01/21/social-
media-a-definition/
5. Conrad, J., Levinson, J., & Gibson, S. (2010). Guerrilla social media marketing. Irvine,
CA: Entrepreneur Press.
6. Deighton, J. K. (2007, Sept 26). Digital Interactivity: Unanticipated Consequences for
Markets, Marketing, and Consumers. Retrieved June 20, 2014 from HBS. edu:
http://www.hbs.edu/faculty/ Publication%20Files/08-017_1903b556-786c49fb-8e95-
ab9976da8b4b.pdf
7. Donath, J. (2004). Sociable Media. Prepared for The Encyclopedia of Human-Computer
Interaction. Print
8. Yu, H.; Hu, Y.; Shi, P. A prediction method of peak time popularity based on twitter
hashtags. IEEE Access 2020, 8, 61453–61461. [CrossRef]
9. 9. Huang, J.; Tang, Y.; Hu, Y.; Li, J.; Hu, C. Predicting the active period of popularity
evolution: A case study on Twitter hashtags. Inf. Sci. 2020, 512, 315–326. [CrossRef]
10. 10. Figueiredo, F.; Jorge, A. Identifying topic relevant hashtags in Twitter streams. Inf.
Sci. 2019, 505, 65–83. [CrossRef]
11. 11. Chowdhury, J.R.; Caragea, C.; Caragea, D. On Identifying Hashtags in Disaster Twitter
Data. In Proceedings of the AAAI Conference on Artificial Intelligence, New York, NY,
USA, 7–12 February 2020; Volume 34, No. 01. pp. 498–506.

36
12. 12. Rho, E.H.R.; Mazmanian, M. Political Hashtags & the Lost Art of Democratic
Discourse. In Proceedings of the 2020 CHI Conference on Human Factors in Computing
Systems, Honolulu, HI, USA, 25–30 April 2020; pp. 1–13.
13. 13. Li, L.; Liu, J.; Sun, Y.; Xu, G.; Yuan, J.; Zhong, L. Unsupervised keyword extraction
from microblog posts via hashtags. J. Web Eng. 2018, 17, 93–120.
14. 14. Simeon, C.; Hamilton, H.J.; Hilderman, R.J. Word segmentation algorithms with
lexical resources for hashtag classification. In Proceedings of the IEEE International
Conference on Data Science and Advanced Analytics (DSAA), Montreal, QC, Canada, 17
October 2016; pp. 743–751.
15. 15. Simeon, C.; Hilderman, R. Evaluating the Effectiveness of Hashtags as Predictors of
the Sentiment of Tweets. In Proceedings of the International Conference on Discovery
Science, Banff, AB, Canada, 4–6 October 2015; Springer: Cham, Switzerland, 2015; pp.
251–265.

37