Sentiment Analysis of Tweets

Using Python
Dr. Ritesh Srivastava1, Bharat Singh2, Choudhary Rishab Kumar2, Prashant Raj2
1. Associate Professor, Department of Computer Science & Engineering
2. Student, Department of Computer Science & Engineering
Galgotias College of Engineering and Technology,
Greater Noida, U.P, India.

Abstract - This paper aims to present an overview of impact of new products in the market can be predicted by
sentimental analysis. Sentimental analysis works in the sentiment analysis.
background of many industries. Today the industries have
become more considerate to their customer’s demands, likes, Generally, there are three types of sentiments in which we can
dislikes, opinions, responses etc. The bright minds of the analyze any data. Positive sentiments, Negative sentiments,
consumer market have realized the power of opinions. The and Neutral sentiments are the types of sentiment analysis.
producers have learned to respect the needs and ideas of their
customers, as they know in this date of neck-to-neck The aim is to extract information from the text messages of the
competitions, their customers have the power to make and user and use it for various purposes like sentiments analysis.
break them. Keeping an eye on the customer response has The model also includes the analysis of emoticons in order to
become a crucial part of production system. Sentiment completely parse the statements the aim is to extract
analysis has come out as a powerful tool to make this process information from the text messages of the user and use it for
easier and much faster. Sentiment analysis involves collection different purposes such as sentiments analysis. In this
of data, pre-processing and polarity detection. component, the data is assigned a sentiment such as positive or
negative and the extent of it by performing data pre-processing
Keywords: Text Mining, Data Mining, Information Extraction, using the SVM algorithm.
Natural Language Processing, Concept Mining.
II. IMPORTANCE AND BACKGROUND
I. INTRODUCTION
Opinions are fundamental to every single act since they're key
Sentiment analysis is that the process of using text influencers of our practices. At whatever point we've to
analysis, statistics, and natural language processing to research choose a choice, we'd like to understand others' thoughts. In
customer sentiment. The most effective companies recognize reality, agencies and institutions dependably want to find out
the sentiment in their customers—what human beings are users’ famous sentiments approximately their gadgets and
pronouncing, how they’re pronouncing it, and what they services.
mean. Customer sentiment is frequently determined in tweets,
comments, reviews, or different locations in which human Clients use distinct sorts of online systems for social
beings point out your brand. Sentiment Analysis is that the engagement along with web-primarily based totally social
area of knowledge of those feelings with software, and it’s a networking sites; for example, Facebook and Twitter. Through
must-recognize for builders and enterprise leaders in the those web-primarily based totally social networks, purchaser
course of a current workplace [1]. engagement occurs progressively.

As with many specific fields, advances in deep gaining This kind of connection offers an interesting open door for
knowledge have delivered sentiment analysis into the advertising knowledge. Individuals of each nationality, sexual
foreground of modern-day algorithms. Today we use statistics, orientation, race, and sophistication utilize the online to share
text analysis, and natural language processing to extract, and encounters and impressions about virtually every feature of
identify the sentiment of words into positive, negative, or their lives. Other than composing messages, blogging, or
neutral categories. leaving remarks on corporate sites, an excellent many
individuals utilize informal organization destinations to log
Sentiment Analysis is generally used for brand monitoring, opinions, express emotion, and uncover insights about their
customer service, market research, and analysis, etc. everyday lives.
In the field of business/marketing to know the specific reason
why a customer is not responding to something (Like provided Individuals compose correspondence on almost anything,
services are meeting with customer’s expectations are not?), together with films, brands, or social exercises. These logs
markets use sentiment analysis. To get an all-over view of flow at some stage in online agencies and are digital
how your product/company is viewing by stakeholders and gatherings wherein consumers remove darkness from and
customers, Sentiment analysis is used. Also, to know the affect others. To the advertiser, these logs provide profound
snippets of insight into purchasers’ behavioral inclinations and
present endless opportunities to seek out client emotions and
recognitions, as they happen without interruption or
incitement. Be that as it may, current explosions in client-
produced content material on social websites are introducing
specific problems in capturing, inspecting, and translating
revealed content material due to the fact records are scattered,
confused, and divided [2].

Opinion investigation may be a method of data mining that

will overcome these difficulties by methodically separating
and dissecting web-based information without causing delays.
With conclusion examination, advertisers are ready to
discover shoppers’ emotions and states of mind continuously,
in spite of the difficulties of data structure and volume. The Fig 1: How to Import Packages [3]
enthusiasm during this study for utilizing sentiment analysis as
an instrument for promoting research instrument is twofold.
Data Set Description: The data is obtained by extracting all
Sentiment analysis critically encourages organizations to work the text messages sent by the subject. This can be taken from
out customers’ likes and dislikes about products and company Twitter. All the tweets posted on Twitter are stored in a
image. In addition, it plays an important role in analyzing data database, we can analyze the sentiments on there.
of industries and organizations to assist them in making The data set will contain tweets in text format and emojis.
business decisions. Data, that is in another format cannot be analyzed.

Using the API of Twitter, we will be able to collect twitter

III. PROPOSED METHODOLOGY dataset for the sentiment analysis.
The aim is to extract information from the text messages of the There are many types of APIs and tool which can be used to
user and use it for sentiments analysis. crawl and collect data:
The model also includes the analysis of emoticons in order to a) Twitter’s Firehose
completely parse the statements the aim is to extract b) Twitter’s Search API
information from the text messages of the user and use it for c) Twitter’s Streaming API
different purposes such as sentiments analysis. In this d) NodeXL
component, the data is assigned a sentiment such as positive or
negative and the extent of it by performing data pre-processing
using the SVM algorithm.

IV. SENTIMENT ANALYSIS WITH PYTHON

So, there are numerous kinds of sentiment evaluation models.
There also are many approaches to do sentiment evaluation.
We’ll give attention to making use of any such methods. That
is growing a bag-of-phrases version of the usage of an SVM. Table 1: Example Tweets[4]
Let’s begin with the Python applications so that it will assist
us to do this.

Packages: There are a lot of python packages are available

which can be used the sentiment analysis. Some of the
available packages are NumPy, Pandas, matplotlib,
wordcloud, etc. We can directly import these packages in our
code while implementing.
Some packages that are used in Sentiment analysis:
NumPy
Pandas Fig 2: Code Snippet to load the DataSet[5]
Wordcloud
TextBlob
Matplotlib Text Cleaning: Initial step of sentiment Analysis techniques is
TensorFlow applied to data in order to reduce the noise of text, reduce
dimensionality, and assist in the improvement of classification Result Analysis:
effectiveness.

There are many types of techniques to pre-process data:

a) Remove numbers
b) Stemming
c) Remove stopwords
d) Part of speech tagging

Fig 5: Result Analysis of Narendramodi’s Twitter Account

V. CONCLUSION
The proposed model takes input from the data set created by
accumulating all the text messages send by the subject. All the
Fig 3: Code Snippet to Clean Data [6] messages may be from different social media platforms such
as Facebook, WhatsApp, etc.

The messages are then pre-processed to obtain the key words

from the data sets. After pre-processing we use probabilistic
language models like n gram. Associating weights to the data
set using TF-Idf increases overall efficiency of classifying
algorithms. The next step is to use the classifying algorithms
to classify the conversations as “normal” or “critical”. [8] First
a supervised algorithm is used which is SVM as it proves to be
highly efficient for such computations and then an
unsupervised algorithm is used which in turn increases the
Table 2: Cleaned Tweet Examples [7] efficiency drastically, in our case we use the KNN algorithm.

Classification of Sentiments: There are different types of Thus, we propose to give a highly efficient method [9] of
sentiment analysis in which it can be performed: finding the sentiment of the person by analysing the text
a) Machine Learning messages and also processing emoticons. Emoticons [10] are
b) Lexicon-Based very common tokens in any text message in the new world,
c) Hybrid therefore we must also focus on efficient ways to analyse
them. We have converted emoticons to textual form for our
computation processes. Thus, this model is a requirement and
a life saviour in the modern world.

Fig 4: Classification of Sentiments

 REFERENCES
[1] https://algorithmia.com/blog/introduction-sentiment-analysis
[2] A. M. Kaplan and M. Haenlein, "Users of the world, unite! The
challenges and opportunities of Social Media," Business horizons, vol. 53,
no. 1, pp. 59-68, 2010.
[3] https://towardsdatascience.com/introduction-to-sentiment-analysis-
f623f7d40bfa
[4] https://towardsdatascience.com/introduction-to-sentiment-analysis-
f623f7d40bfa
[5] https://towardsdatascience.com/introduction-to-sentiment-analysis-
f623f7d40bfa
[6] https://towardsdatascience.com/introduction-to-sentiment-analysis-
f623f7d40bfa
[7] https://towardsdatascience.com/introduction-to-sentiment-analysis-
f623f7d40bfa
[8] Giles, C. L. Bollacker, K. D. And Lawrence, S. 1998. Cite Seer: an
automatic citation indexing system, In Digital Libraries 98 -The Third
ACM Conference on Digital Libraries, pp. 89-98.
[9] Garfield, E. and Well jams-Dor of, A. 1992. Citation data: their use as
quantitative indicators for science and technology evaluation and policy-
making, Science & Public Policy, vol. 19, no. 5, pp. 321-327.
[10] I. H. Witten, K. J. Don, M. Dewsnip, and V. Tablan, ‘‘Text mining in a
digital library,’’ Int. J. Digit. Libraries, vol. 4, no. 1, pp. 56–59, 2004.