Sentiment-Analysis---NLPLakshitha A
The analysis began with Exploratory Data Analysis (EDA) to uncover patterns in hate speech and visualize word distributions using libraries like matplotlib and seaborn. Several machine learning models were trained and evaluated using scikit-learn, with Logistic Regression emerging as the best model, achieving an impressive accuracy of 95.00%.
This project underscores the potential of AI and NLP to foster safer digital environments by automating the detection of harmful content. Future enhancements could include integrating real-time tweet analysis with the Twitter API, experimenting with advanced models like XGBoost or LSTMs, and refining preprocessing steps with techniques such as stemming and lemmatization.
What's included
Twitter-Sentiment-Analysis---NLP
This project focuses on detecting hate speech in tweets, specifically targeting racist and sexist sentiments. Using a labeled dataset of 31,962 tweets, the goal was to classify tweets as hate speech or not. The project employed Natural Language Processing (NLP) techniques for data preprocessing, including tokenization, stop-word removal, and converting text to numerical features using TF-IDF vectorization.
Lakshitha's other services
Starting at$40
Duration6 weeks
Tags
MATLAB
Microsoft Excel
pandas
Tableau
Data Analyst
Data Modelling Analyst
Data Scientist
Service provided by
Lakshitha A Madurai, India
Sentiment-Analysis---NLPLakshitha A
Starting at$40
Duration6 weeks
Tags
MATLAB
Microsoft Excel
pandas
Tableau
Data Analyst
Data Modelling Analyst
Data Scientist
The analysis began with Exploratory Data Analysis (EDA) to uncover patterns in hate speech and visualize word distributions using libraries like matplotlib and seaborn. Several machine learning models were trained and evaluated using scikit-learn, with Logistic Regression emerging as the best model, achieving an impressive accuracy of 95.00%.
This project underscores the potential of AI and NLP to foster safer digital environments by automating the detection of harmful content. Future enhancements could include integrating real-time tweet analysis with the Twitter API, experimenting with advanced models like XGBoost or LSTMs, and refining preprocessing steps with techniques such as stemming and lemmatization.
What's included
Twitter-Sentiment-Analysis---NLP
This project focuses on detecting hate speech in tweets, specifically targeting racist and sexist sentiments. Using a labeled dataset of 31,962 tweets, the goal was to classify tweets as hate speech or not. The project employed Natural Language Processing (NLP) techniques for data preprocessing, including tokenization, stop-word removal, and converting text to numerical features using TF-IDF vectorization.
Lakshitha's other services
$40