Twitter Sentiment Analysis

Introduction

Nowadays text mining and sentiment analysis are sensational topics in the area of research. Twitter, a social media website which allows different ways of demonstration opinions and establish communication among the users, all over the world. Since Twitter is very distinctive so it’s difficult to assemble data for sentiment classification. , data is first preprocessed before analysis, to enhance the tweet. After tweets are processed, different information and features are produced from them.[1]

Twitter, a social networking service where users can post and communicate with messages basically called “Tweets' in its terminology. ,but only registered users can share their views in the form of likes, comments , posts , retweets and unregistered users can just read them . Twitter was called into existence in March 2006 by Jack Dorsey , Biz Stone, Noah Glass and Evan Williams and instigated in identical year July. Earlier, more than 120 million users posted 400 million tweets per day, and besides service touched a standard of 2 billion search questions everyday.[2]

Twitter Sentimental Analysis focuses on the mood during which the user posted the tweet. It tells the polarity of text, categorizes the tweet into three branches of expression that are positive,negative or neutral. For analysis tweet must be limited under 140 characters. User needs to convey their message within the character limit. There are a large number of tweets in a day there are a large amount of tweets posted by users because there are a large number of users on this platform so it becomes difficult to analyse what type of tweet was posted by the user. In this project Twitter Sentimental Analysis, we will categorize the tweets which are positive, negative or neutral. The primary task in sentiment analysis is to classify the polarity of a given text at the document, sentence. Advanced sentiment classification looks for emotional states such as whether the sentence shows 'angry', 'sad', or 'happy' emotion of the user.

Currently there are three main approaches for sentiment analysis which are statistical methods, hybrid approaches, and knowledge based approaches.[3]

Save your time!
We can take care of your essay

• Proper editing and formatting
• Free revision, title page, and bibliography
• Flexible prices and money-back guarantee

Place Order

For this technique Text mining and sentimental analysis are used they are used to analyze unstructured tweet text to get positive and negative polarity about this text. Also, tweet frequency analysis is done to view the changing trend in public opinion across a time interval of 9 days’ tweet text data. It is found that a large number of people have this attitude towards this incident by using 2-3 hashtag with overall data.

So why is Twitter Sentimental analysis important?

It helps to Analyze thousands of tweets which are mentioning your brand or your business. Also used for real time analysis: It is also used to monitor Twitter sentiment analysis is essential to observe sudden shifts in customer moods, detecting if complaints are increasing, and for taking action accordingly. With sentiment analysis, you can also monitor brand mentions and gain actionable insights.[4].

One important step in sentiment analysis is preparing the dataset because the language that is generally used in the tweets is not a level language, it is mostly some short forms or some fancy words . Most of the times the spelling or the utilization of the words used may vary a lot from people to people. As it is a superintended learning technique, so it requires a training set to tell the polarity of the text. Data Preprocessing becomes very necessary because of the use of patter, URL removal and mis spelled words. To solve decipher problems because of patter words, a dictionary is retained, which analyses the words and after that returns the word with its similar or identical meaning.

SENTIMENT CATEGORIZATION - Classification is done after performing above steps using the machine learning algorithm which is Naive Bayes, It Keeps Up Vector Machine, Maximum Entropy and Troupe Classifiers. Categorization Techniques Normally, there are a different kind of classifier required for categorization of text in twitter analysis. Naive Bayes Classifier The algorithm in the naive Bayes classifier requires all characteristics of the feature vector. It tells the conditional probability. Sentiment Analysis can be done so productively on Twitter because of the existence of predefined properties like emotional keyword, a number of positive and negative hashtag, the number of keywords which are positive or negative; it also observes the emotional keyword used and emojis used. The relationship existing between the characteristics is not regarded in this classifier for categorization. [5]

References

1. Nagamanjula, R., & Pethalakshmi, A. (2020). A novel framework based on bi-objective optimization and LAN2FIS for Twitter sentiment analysis. Social Network Analysis and Mining, 10(1). doi:10.1007/s13278-020-00648-5
2. Wikipedia contributors. (2020, October 26). Sentiment analysis. Retrieved November 6, 2020, from Wikipedia, The Free Encyclopedia website: https://en.wikipedia.org/w/index.php?title=Sentiment_analysis&oldid=985616558
3. Mishra, P. (2020). Twitter Sentimental Analysis. International Journal for Research in Applied Science and Engineering Technology, 8(5), 2476–2478.
4. Karumanchi, B. (2020). An unsupervised clustering approach for twitter sentimental analysis: A case study for George Floyd incident. International Journal of Computer Trends and Technology, 68(6), 46–50.
5. Singh, S. S., Dwarakanath, D., Santhoshini, & Mary, J. (2020). Twitter Sentimental Analysis. Journal of Advanced Research in Dynamical and Control Systems, 12(05-SPECIAL), 868–873.