Scribd Logo
Upload

Welcome to Scribd!

  • 631 views

    IMDb Movie Review Sentiment Analysis

    Uploaded by

    Sagar Bhutada
    This document presents an IMDb web scraping and data analysis project. The project scrapes over 3,500 movie records from IMDb, including details like title, genre, year, ratings, budget, and reviews. The data is stored in a MySQL database and analyzed using Python. Sentiment analysis is performed on the reviews to classify them as positive or negative. A support vector machine model is built and achieves 79.61% accuracy in predicting review sentiment. Several data visualizations are created using Seaborn to provide insights like top rated movies, high budget movies, and distributions of ratings and budgets. Challenges addressed include scraping accurate budget data and converting reviews to numeric vectors for modeling.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd

    IMDb Movie Review Sentiment Analysis

    Uploaded by

    Sagar Bhutada
    631 views18 pages
    This document presents an IMDb web scraping and data analysis project. The project scrapes over 3,500 movie records from IMDb, including details like title, genre, year, ratings, budget, and reviews. The data is stored in a MySQL database and analyzed using Python. Sentiment analysis is performed on the reviews to classify them as positive or negative. A support vector machine model is built and achieves 79.61% accuracy in predicting review sentiment. Several data visualizations are created using Seaborn to provide insights like top rated movies, high budget movies, and distributions of ratings and budgets. Challenges addressed include scraping accurate budget data and converting reviews to numeric vectors for modeling.

    Original Description:

    to perform sentiment analysis on review and analysis on the given data

    Original Title

    IMDb movie review sentiment analysis

    Copyright

    © © All Rights Reserved

    Available Formats

    PPTX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document presents an IMDb web scraping and data analysis project. The project scrapes over 3,500 movie records from IMDb, including details like title, genre, year, ratings, budget, and reviews. The data is stored in a MySQL database and analyzed using Python. Sentiment analysis is performed on the reviews to classify them as positive or negative. A support vector machine model is built and achieves 79.61% accuracy in predicting review sentiment. Several data visualizations are created using Seaborn to provide insights like top rated movies, high budget movies, and distributions of ratings and budgets. Challenges addressed include scraping accurate budget data and converting reviews to numeric vectors for modeling.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    631 views18 pages

    IMDb Movie Review Sentiment Analysis

    Uploaded by

    Sagar Bhutada
    This document presents an IMDb web scraping and data analysis project. The project scrapes over 3,500 movie records from IMDb, including details like title, genre, year, ratings, budget, and reviews. The data is stored in a MySQL database and analyzed using Python. Sentiment analysis is performed on the reviews to classify them as positive or negative. A support vector machine model is built and achieves 79.61% accuracy in predicting review sentiment. Several data visualizations are created using Seaborn to provide insights like top rated movies, high budget movies, and distributions of ratings and budgets. Challenges addressed include scraping accurate budget data and converting reviews to numeric vectors for modeling.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    of 18

    IMDb Web Scraping and Data Analysis

    Presented By:-
    Shehzada Alam
    Sameer Pophali
    Shreyas Wankhede
    Sagar Bhutada
    Shehnaaz Shareen
    Agenda
    • Problem Statement
    • Proposed Solution
    • Solution Details
    • Model Building
    • Challenges
    • Data Analysis & Visualization.
    • Conclusion
    Problem Statement
    • The Internet Movie Database (IMDb) is one of the world’s most popular sources for movie, TV and celebrity
    content with more than 100 million unique visitors per month.

    • IMDb has huge collection of movies database that includes various details of movies along with different ratings
    and user reviews.

    • This movie reviews affects everyone from audience, film critics to the production company.

    • Idea of our project is to scarp the data from IMDb and form an analysis that will help data analyst or production
    company to decide how they are going to proceed with making a new movie, second is to form a model to predict
    what are the sentiments of movies based on user reviews.

    • Dataset: 3500+ records and has 10 columns.


    Proposed System:
    Solution Details:
    • Extracting all the details of movies like IMDBID,Title,Genre, Year, Audience rating, Critics rating, Budget and Reviews
    using 'OMDB API' and webscrapping.

    • Established the Database Connection as we are storing this movie data in MySQL.

    • While searching the movie detail if the entry is not present then fetch the detail from imdb through webscraping and
    with API, insert the record in database and display result back to the user.

    • For analysis extract the data from database into dataframe and visualize the data to get some insights.
    • “Sentiment analysis is an important research area that identifies the people’s sentiments, opinions and emotions
    underlying a text.”

    • Sentiment Analysis based on User Reviews and created a new column(polarity) which includes this Labels (Positive and
    Negative).

    • We have used “Unsupervised lexicon based method”, which are dictionaries or vocabularies of polar words specially
    constructed for sentiment classification task.

    • The system uses VADER (Valence Aware Dictionary and sEntiment Reasoner) based lexicon method for sentiment
    analysis that not only tells about the positive,negative, neutral and compound score between -1 to +1 but gives positive
    or negative sentiment of reviews based on this score.

    • To extract this sentiments we have computed the polarity of the given reviews (whether the text is positive or
    negative).
    Model Building
    • In order to make sense to our machine learning algorithm we have converted each review to a numeric
    representation which is called 'Vectorization'.

    • The system uses TF-IDF Vectorizer (Term Frequency-Inverse document frequency) that transforms a count
    matrix to a normalized frequency representation in float.

    • Splitting the movie data into Train and Test set (80-20 ratio).

    • SVM: The objective of a Linear SVC (Support Vector Classifier) is to fit the data, returning a best fit hyperplane
    that divides or classifies the data.
    Accuracy= 79.61%
    Precision Recall f1-score support
    Negative 0.77 0.82 0.79 371
    Positive 0.83 0.77 0.80 404
    Total 0.80 0.80 0.80 775
    Django Framework
    Frontend
    Challenges
    • Using omdb API we were not able to fetch Budget and User Reviews hence we have scrapped the data
    from imdb website using imdb id.

    • While fetching 'Budget' from imdb website the amount were present in different currency format so we
    have converted the currencies in USD by using CurrencyConverter package.

    • After getting polarity for all the user reviews, this reviews was needed to be converted in numerical
    representation to fit the classification model.

    • While rendering seaborn graphs on Django Framework we were getting several response errors and
    were not able to display the plots.
    Data Analysis and Visualization:
    1) Top 10 Rated Movies:
    2) Top 10 High Budget Movies:
    Critic Vs Audience Rating:
    Distribution of Critic or Audience Rating:

    Critic Rating Audience Rating


    Stack Distribution of Movie Budget:
    BARPLOT:
    BOXPLOT:
    Thank You

    You might also like