Answer To: CONDUCT A PERSONAL RESEARCH PROJECT BY APPLYING DATA MINING AND MODELING TECHNIQUES Individual...
Aditi answered on Jul 21 2022
Twitter sentiment analysis
[TWITTER SENTIMENT ANALYSIS]
Final Project Report
Twitter Sentiment Analysis
Abstract
[TWITTER SENTIMENT ANALYSIS]
May 28, 2014
Over the course of the last decade, humankind has seen an exponential increase in the usage of online resources, particularly social media and websites that facilitate microblogging like Twitter. Numerous businesses and groups have discovered the wealth of marketing information that can be gleaned from these sources. This research focuses on the implementation of machine learning techniques in order to extract the sentiment of an audience in relation to a well-known television show. Comparing several machine learning algorithms with regard to the undertaking of sentiment categorization was one of the primary focuses of this investigation. The most important discovery was that, out of all of the classification techniques that were tested, it was discovered that the Random Forest model offered the greatest accuracy of classification for this particular area. Based on the findings of this investigation, it is possible to draw the conclusion that the machine learning - powered approaches that were provided are methods for sentiment classification that are both effective and practical.
iii
Table of Contents
Abstract ii
Section 1.a: Introduction 1
Study Rationale 1
Research Objectives 1
Significance of Research 1
The innovativeness of the project 2
Structure of the Research project/report 2
Section 1.b: Literature Review 3
Hypothesis 3
Conceptual Modeling 3
Section 2: Research Methodology 3
Overview of the Methodology 3
Naïve Bayes 4
Random Forests 4
Data collection process and techniques 5
Platform 5
Data Collection 5
Training Data 6
Data cleaning process 7
Section 3: Data Analysis 8
Overview of Data Analysis 8
Python 8
SQL 9
Discussions 9
Section 4: Research Implications and Conclusion 9
Practical Recommendations 9
Theoretical Recommendations 9
Study Limitations 10
Conclusion 10
References 10
Section 1.a: Introduction
This section will cover the overall aim of the project, the motivation behind it, and any contribution to the knowledge base that has been added.
Study Rationale
This research aimed to classify a TV broadcast's emotion. Twitter was mined for audience opinions on "Supernatural." This kind of sentiment analysis aims to determine how people feel about a TV broadcast. Twitter information is beneficial or bad. The classified data will be analyzed to determine what percentage of the sample population belongs to each category.
This article evaluates multiple machine learning algorithms for evaluating Twitter sentiment.
Research Objectives
Social media and microblogging sites like Twitter, Facebook, and YouTube have skyrocketed in popularity in the recent decade. Businesses and organizations perceive these sources as a marketing gold mine. Interviews, questionnaires, and surveys were traditional customer feedback techniques. Contextual restrictions and poorly prepared surveys made traditional procedures time-consuming and expensive for enterprises. Natural language processing and sentiment analysis are improving marketing decisions and product feedback. Every day, gigabytes of consumer-related data, largely in unstructured language, are released online. With Moore's law (1965) and frameworks like Hadoop, processing huge datasets is possible. IBM, which is developing its NLP supercomputer Watson, and Google, which recently bought deep mind technology, are spending substantial resources to this sector. Data analytics and search engines will benefit from more research so robots can "understand" language.
Significance of Research
There are many businesses and organizations that place a high value on client feedback about their products and services. A business may use the results of this kind of study to better understand its customers, develop better goods, communicate with its target market, and measure the success of its marketing initiatives. These insights help businesses collect crucial user input for use in developing the next iteration of their product.
The findings of the sentiment analysis conducted for this purpose will provide the show's creators insight into how each episode is being received by the audience. As viewers share their thoughts on the show before, during, and after it airs, this data becomes more important. An application like this might analyze the sentiment in real time, providing producers with instantaneous feedback on how the show is being received in the eyes of its viewers. One may see a future version of this software using clustering techniques to shed light on certain situations or people. Academics thought that no new insights into natural language processing or sentiment analysis had been discovered. This was to be anticipated, given the advanced nature of the subject matter covered in this course (level 8 on the national framework) and the limited time allotted for the project. However, this kind of application may help small and medium-sized businesses acquire insights from their data without having to devote a big budget to the cause. Since this paper is one of the few in the twitter sentiment analysis sector that evaluate many machine learning classification algorithms and produce such a high-quality final model, it may also be used as a reference by other researchers.
The innovativeness of the project
Twitter sentiment research may warn you to possible problems like unsatisfied customers or negative reviews before they become serious concerns. Twitter sentiment analysis opens up new opportunities. Real-time Twitter sentiment analysis boosts digital marketing by revealing messages' underlying tone.
Social media-obsessed
Online reputation is crucial. If a company doesn't reply swiftly to a poor social media review, it might lose money.
Client assistance
Customer care specialists must have a Twitter presence. They must address customers' inquiries quickly.
Twitter users provide crucial market input. It expresses people's ideas, feelings, observations, and perspectives on many topics.
Twitter sentiment analysis may be used to monitor particular phrases and topics to detect customer preferences. If you want a new product launch to be successful, you must know what customers value and how they will act.
Structure of the Research project/report
The research is structured into different components where we have...