https://github.com/adithivs/prodigy_ds_04
https://github.com/adithivs/prodigy_ds_04
jupyter python3 sentiment-analysis visualization
Last synced: 4 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/adithivs/prodigy_ds_04
- Owner: AdithiVS
- License: bsd-2-clause
- Created: 2024-06-25T08:03:57.000Z (almost 2 years ago)
- Default Branch: main
- Last Pushed: 2024-06-28T14:47:21.000Z (almost 2 years ago)
- Last Synced: 2025-01-12T06:07:37.712Z (over 1 year ago)
- Topics: jupyter, python3, sentiment-analysis, visualization
- Language: Jupyter Notebook
- Homepage:
- Size: 3.97 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# PRODIGY_DS_04
# Twitter Sentiment Analysis
## Task Description
Analyzing Twitter tweets' attitude toward a specific entity is the aim of this work. Sorting each message's attitude into one of three categories—Positive, Negative, or Neutral—is the work at hand. Neutral messages are those that don't pertain to the designated entity. This will help us to learn more about how the Twitter community feels and expresses opinions on specific entities by carrying out this sentiment analysis.
## About Dataset
This dataset focuses on entity-level sentiment analysis on Twitter data. The objective is to determine the sentiment of a message towards a specific entity. The dataset comprises three sentiment classes: Positive, Negative, and Neutral. Messages that are deemed irrelevant to the entity are categorized as Neutral. The task involves assessing the sentiment of messages concerning the specified entity across various social media interactions.
## Findings from the Data After Sentiment Analysis
### Overall Sentiment Distribution
The bar plot shows the count of each sentiment (Negative, Neutral, Positive) across the entire dataset. This helps in understanding the general sentiment trend in the dataset.
### Sentiment Distribution by Topic
The table and the stacked bar plot show how sentiments are distributed across different topics. This helps in identifying which topics have more positive, negative, or neutral sentiments. For example, topics like "AssassinsCreed" have a high count of positive sentiments, while "Battlefield" has a significant number of negative sentiments.
### Sentiment Distribution of a Specific Topic
We can find sentiment distribution of any topic. For example, here we have analyzed Google and Facebook:
- **Google**: 15.3% were positive and 35.8% were neutral.
- **Facebook**: 6.7% is positive and 33.7% is neutral.
### Word Cloud Insights
- **Google**: The word cloud for "Google" likely highlights common terms associated with the company, such as "search", "engine", "ads", "YouTube", "Android", etc. This visualization can help identify key themes and topics of discussion related to Google.
- **Facebook**: The word cloud for "Facebook" likely highlights terms such as "social", "media", "friends", "posts", "privacy", etc. This visualization can help identify key themes and topics of discussion related to Facebook.
## Contact Information
- Adithi Vellengara(LinkedIn)
- Email 📧: adithivs06@gmail.com
## Conclusion
This sentiment analysis helps in understanding the Twitter community's feelings and opinions on specific entities. By analyzing the sentiment distribution and key terms associated with different topics, we can gain valuable insights into public perception and discussions on social media.