Sentiment Analysis.pptx

Table of Contents
2
Problem Definition
Introduction
Motivation
Literature Survey
Implementation
I
II
III
IV
V
VI
Algorithm taken to implement
VII Conclusion
VIII References

2
PROBLEM DEFINITION
Sentiment Analysis and Classification of Tweets
Using Classification Algorithms like Naive Bayes
and K-NN Algorithm.

2
In recent times, people have started using social media sites more often and it has become an
important aspect to analyze and find what they think about the content on social media and
what reactions they give. Data classification is the process of categorizing analyzed data to
make it more effective and productive. The classification's goal is to accurately predict the
target class for each and every case in the data. A classifier is basically an algorithm that is
specifically designed to implement classification. The data that is collected from the users
should be analysed and it should be properly processed and used for the identification of the
sentiments of the users. . The data set that we get from the users is very large and not all of it is
useful, proper selection and extraction of data needs to be done in order to separate out the
useful data.
I
INTRODUCTION
3

2
So, text mining is the process that is used for extraction of the useful data. It helps in getting the
exact information and meaningful data from the text. For applying the data mining algorithms
some processes need to be carried out, such as natural language processing or information
retrieval methods, or some pre-processing of text. .We will be using two classifiers for the
extraction of sentiments and thoughts of people based on what they share on Twitter and how
they respond to other people’s tweets and classify their tweets into categories. By comparing
the results of all two classifiers we will be able to find out which classifier gives the most
accurate result in terms of precision and accuracy
I
INTRODUCTION
4

2
MOTIVATION
For new technologies in this decade, we need a solution that
can create the perfect model to solve that particular problem.
The sentiments of the people are difficult to understand when
it comes to social media and hence it needs to be analyzed
and processed properly and accurately.

2
No. Techniques Summary limitations
1
Naïve Bayes algorithm
Naïve Bayes algorithm is
a classifier learning algorithm. It's
supported by the Bayes theorem and
used
for finding classification of datasets.
This algorithm faces the ‘zero-
frequency problem’ where it assigns
zero probability to a variable whose
category in the test data set wasn’t
available in the training dataset
2
KNN-precision
K-NN algorithm is
K-Nearest Neighbor and it checks for
the similarity between the new data
and
available data and puts our new case in
a category that best fits.
Accuracy
depends on the type of the data.
With
large data, the prediction stage might
become slow.
LITERATURE SURVEY

2
ALGORITHM
The two algorithms we have used to classify the tweets are:
• KNN Algorithm
• Naive Bayes Algorithm
KNN Algorithm:
K-NN algorithm is K-Nearest Neighbour and it checks for the similarity between the new
data and available data and puts our new case in a category that best fits. It mainly works on
the principle of similarity. It is a non-parametric algorithm, it learns from the dataset while it
performs actions on it and not immediately. KNN is mainly used for differentiating the data
into categories and for the identification of the class and category of a dataset.

2
ALGORITHM
Bayes Algorithm:
Naïve Bayes algorithm is a classifier learning algorithm. It's supported by the Bayes theorem
and used for finding the classification of datasets. Bayes theorem is also conjointly referred
to as Bayes Rule or Bayes law. It's used to calculate the chance of a specific dataset.
The formula for Bayes theorem is given as:
P(A|B) = P(B|A)P(A)/P(B)
Naïve Bayes formula is especially employed in text classification that features a high-
dimensional dataset. It's a probabilistic classifier, meaning it's used to predict on the premise
of the probability of an object. It's employed in Text classification like Spam filtering and
Sentiment analysis.

2
I
REFERENCES
4
[1] Akrati Saxena, Harita Reddy, Pratishtha Saxena, "Introduction to Sentiment Analysis Covering Basics,
Tools, Evaluation Metrics, Challenges, and Applications", Principles of Social Networking, vol.246, pp.249,
2022.
[2] L. Tan, J. Na, Y. Theng, K. Chang, “Sentence-level Sentiment polarity classification using a
linguistic approach”, Digital Libraries: For Cultural Heritage, Knowledge Dissemination, and Future
Creation (2011) 77-87.
[3] Pratama BY, Sarno R., “Personality classification based on Twitter text using Naive Bayes, KNN and
SVM”. Proceedings of 2015 International Conference on Data and Software Engineering,ICODSE 2015;
2016. pp. 170–174. https://doi.org/10.1109/ICODSE.2015.7436992.
[4] R. Feldman, “Techniques and applications for
sentiment analysis”, Commun ACM, 56 (2013), pp. 82-
89.
[5] B. Liu, “Sentiment analysis and opinion
mining”, Synth Lect Human Lang Technol (2012)
[6] Agarwal, B. Xie, I. Vovsha, O. Rambow, R.
Passonneau, “Sentiment Analysis of Twitter Data", In
Proceedings of the ACL 2011Workshop on Languages in
Social Media,2011 , pp. 30-38.

2
I
CONCLUSION
4
In this study, we attempted and classified the data according to the
sentiments of the Twitter posts with the help of two algorithms- Naïve
Bayes and K-NN algorithm. The result after performance gives that
the K-NN algorithm has a greater accuracy percentage which is
85.667% while Naïve Bayes gives an accuracy precision of 80% hence
we can conclude that K-NN gives more precise and accurate results
and it should be used for precise prediction of the sentiments of the
users.

Sentiment Analysis.pptx

More Related Content

Similar to Sentiment Analysis.pptx (20)

Recently uploaded (20)

Sentiment Analysis.pptx