UAS - Pengolahan Bahasa Alami dan Text Mining ( Analisis Sentimen )

Summary

This presentation discusses a sentiment analysis project focused on movie reviews. The team imports necessary libraries and cleans the dataset by removing stop words. The dataset includes sentiment labels and review text, which is processed into word clouds to visualize common terms in positive and negative reviews. The team applies logistic regression to classify sentiments, achieving 80% accuracy. The project demonstrates the effectiveness of sentiment analysis in categorizing movie opinions as positive or negative, providing valuable insights into the power of text data analysis and machine learning techniques.

Takeaways

• 😀 The group is presenting their sentiment analysis project on movie reviews.
• 😀 The team consists of three members: Indra Asikaloka, Jilan Azzahra Salsabila, and Miftahul Jannah.
• 😀 The sentiment analysis process began with importing necessary libraries like pandas and other required dependencies.
• 😀 They also imported functions and stopwords to clean the dataset and prepare it for analysis.
• 😀 The team focused on analyzing a sample dataset containing columns like ID, sentiment, and review text.
• 😀 The dataset was cleaned by removing stopwords and unnecessary symbols from the reviews.
• 😀 Sentiment distribution was calculated, showing 100 negative sentiments and only 1 positive sentiment in the sample.
• 😀 A word cloud visualization was created to display the most frequent terms in negative reviews.
• 😀 Another word cloud was generated to display the most frequent terms in positive reviews.
• 😀 The team transformed the review text into vectors and used logistic regression for model training and evaluation.
• 😀 The model achieved an accuracy of 0.8, indicating a good level of performance in predicting sentiment.

Q & A

What is the main topic of the presentation?

The main topic of the presentation is sentiment analysis of movie reviews.

Who are the members of the presenting group?

The presenting group consists of Indra Asikaloka, Jilan Azzahra Salsabila, and Miftahul Jannah.

What libraries and tools are imported for sentiment analysis in the presentation?

The libraries and tools mentioned for sentiment analysis include pandas, and other necessary functions for text processing and stop words removal.

What are the key components of the dataset used in the analysis?

The dataset contains three key components: ID, sentiment, and text. The sentiment can be either positive or negative.

How is the data preprocessed before analysis?

The dataset is preprocessed by cleaning the review column and removing stop words to prepare the data for sentiment analysis.

What sentiment distribution was observed in the dataset?

The sentiment distribution shows that there are 100 negative reviews and only 1 positive review in the analyzed dataset.

How is sentiment visualization presented in the analysis?

Sentiment visualization is presented through word clouds, showing the most frequently occurring words for both negative and positive sentiments.

What machine learning technique is used to predict sentiment?

Logistic regression is used to train the model and predict sentiment in the dataset.

What accuracy was achieved by the logistic regression model?

The logistic regression model achieved an accuracy of 80% in predicting sentiment.

How does the presentation conclude?

The presentation concludes with a polite closing, asking for forgiveness and offering thanks, ending with the greeting 'Assalamualaikum warahmatullahi wabarakatuh'.