SENTIMENT ANALYSIS OF ENGLISH TWEETS USING DATA MINING E-Book

Dr. Gaurav Gupta

• Herausgeber: BookRix
• Kategorie: Fachliteratur
• Sprache: Englisch

Veröffentlichungsjahr: 2018

Beschreibung

Due to the popularity of internet it becomes very easy for people to share their views over social networking websites. Most popular website among them is twitter. Twitter is a widely used social networking website that is used by the numerous people to give their opinion regarding a particular topic or product. So, today it becomes necessary to analyze the tweet of the people. The process to analyze and interpret the tweets is known as sentiment analysis. The main motive of this project is to identify how the tweets on the social networking website are used to identify the opinion of people regarding the particular product or policy. Twitter is a online website that allows the user to post the status of maximum 140 characters. Twitter has over 200 million registered users and 100 million active users [34]. So it comes to be a great source of valuable information. This project aims to develop a better way for sentiment analysis which is nothing a simple way to classify the tweets into positive, negative or neutral. The result of the sentiment analysis can be used by various organizations. Sentiment analysis can be used for forecasting the stock exchange, used to predict the popularity of any product in market, or used to predict the result of elections based on the public views on the social sites. The main motive of project is to develop a better way to accurately classify the unknown tweets according to their content.

Leseprobe

Dr. Gaurav Gupta, Dr. Gurjit Singh Bhathal

SENTIMENT ANALYSIS OF ENGLISH TWEETS USING DATA MINING

Data Mining, Sentiment Analysis

BookRix GmbH & Co. KG81371 Munich

Table of Contents

Abstract

Table of Contents

List of Figures

List of Tables

Chapter 1 INTRODUCTION

1.1 Introduction to Data Mining

1.1.1 Process of Data Mining

1.1.2 Applications of Data Mining

1.1.3 Data Mining Hierarchical Model

1.2 Introduction to Sentiment Analysis

1.2.1 Components of Sentiment Analysis

1.2.2 Level of Sentiment Analysis       

1.2.3 Classification of Sentiment Analysis

1.2.4 Techniques for sentiment Classification

1.2.5 Application Areas

Chapter 2 SURVEY OF LITERATURE

2.1 Introduction

2.2 Related work

2.3 Summary

Chapter 3 METHODOLOGY

3.1 Methodology

3.1.1 Create Dictionary

3.1.2 Tweets Collection

3.1.3 Data Pre-processing

3.1.3.1 Filtering

3.1.3.2 Twitter slang removal

3.1.3.3 Stop words removal

3.1.3.4 Negation Handling

3.1.3.5 Stemming

3.1.3.6 Example for tweets pre-processing

3.1.3.7 Calculating sentiment score

3.2 Algorithm for sentiment Analysis

Chapter 4 IMPLEMENTATION

4.1 Netbeans IDE Interface             

4.2 Main window                            

4.3 Dictionary Creation                    

4.3.1 Positive words dictionary         

4.3.2 Negative words dictionary       

4.4 Slang words table                     

4.5 Stop words table                       

4.6 Tweets dataset                         

4.6.1 IPhone tweets table                

4.6.2 Cricket tweets table                

4.6.3 Badminton tweets table          

4.6.4 Bahuballi2 tweets table           

4.6.5Qismat Punjabi song tweets table

4.6.6Ishqbaaz Hindi serial tweets        

Chapter 5 RESULTS & DISCUSSIONS

5.1 Results for IPhone dataset               

5.2 Results for Bahuballi2 movie dataset 

5.3 Results for Cricket dataset               

5.4 Results for Badminton dataset          

5.5 Results for Ishqbaaz dataset            

5.6 Results for Qismat song dataset       

5.7 Accuracy comparison of different datasets

5.8 Detail of 6 datasets                                

Chapter 6 CONCLUSION & FUTURE SCOPE

6.1 Conclusion     

6.2 Challenges     

6.3 Future Scope  

List of Figures

Figure 1.1:     Process of Data Mining

Figure 1.2:     Data mining hierarchical model

Figure 1.3:     Components of sentiment analysis

Figure 1.4:     Positive, Neutral & Negative sentiment

Figure 3.1:     Architecture of proposed system

Figure 3.2:     Flow chart of the system

Figure 4.1:     Netbeans IDE Interface

Figure 4.2:     Main executable window

Figure 4.3:     Positive words table

Figure 4.4:     Negative words table

Figure 4.5:     Slang words table

Figure 4.6:     Stop words table

Figure 4.7:     Sentiment140 tool

Figure 4.8:     Sentiment140 tool after login to twitter

Figure 4.9:     IPhone tweets table

Figure 4.10:   Cricket tweets table

Figure 4.11:   Badminton tweets table

Figure 4.12:   Bahuballi2 movie tweets

Figure 4.13:   Qismat song tweets

Figure 4.14:   Ishqbaaz serial tweets

Figure 5.1:     Result of IPhone tweets

Figure 5.2:     Pie chart for IPhone tweets

Figure 5.3:     Results of Bahuballi2 movie tweets

Figure 5.4:     Pie chart for Bahuballi2 movie tweets

Figure 5.5:     Result of Cricket tweets

Figure 5.6:     Pie chart for cricket tweets

Figure 5.7:     Result of Badminton tweets

Figure 5.8:     Pie chart for Badminton tweets

Figure 5.9:     Results for Ishqbaaz serial

Figure 5.10:   Pie chart of Ishqbaaz serial tweets

Figure 5.11:   Results of Qismat song tweets

Figure 5.12:   Pie chart for Qismat song tweets

Figure 5.13:   Bar chart showing accuracy of different datasets

Figure 5.14:   Graphical representation of results

List of Tables

Table 2.1:        Summary of Literature Review

Table 3.1:        Database table

Table 3.2:        Positive words table

Table 3.3:        Negative words table

Table 3.4:        IPhone sentiment score database table

Table 3.5:        Data Filtering

Table 3.6:        Slang removal

Table 3.7:        Stemming

Table 3.8:        Example for tweets pre-processing

Table 5.1:        Result table

Chapter-1

INTRODUCTION

Introduction to Data Mining

Data mining is the exploration phases of the "information detection in files" a procedure for determining designs in huge data collections including approaches at the connection of simulated intelligence, machine learning and record systems. The complete objective of the data mining procedure is to mine information from data collections and convert it into a reasonable configuration for additional use. Data mining is a prevailing new skill with excessive prospective to support corporations emphasis on the more essential material in their data stores. Data mining tools forecast upcoming tendencies and performances, permitting industries to create information focused judgments. Data mining tools can replyexpertinterrogations that usually were too time killing to decide. They wash records for unseen patterns, discovery analytical info that specialists may omit as it deceits exterior to their hopes.

The word is a contradiction, as the aim is the drawing out of designs and information from great quantity of data, not the withdrawal of data itself. Data mining methods have progressively been considered, exclusively in their use in actual global databases.

Data mining is a normal progress of the improved use of electronic stores to collect data and deliver replies to commercial specialist.