Lucas Washor - Sentiment Analysis and Film Scripts
Define Sentiment Analysis:
Sentiment Analysis is a computational method that assesses the emotional tone or polarity (positive, negative, neutral) in text by analyzing words and phrases. It transforms raw text into sentiment scores—typically as numerical values for positive, negative, or neutral tones—which allow us to quantify emotions expressed in a body of work. This scoring helps reveal emotional and cultural perspectives in narratives, as well as patterns in how themes are emotionally framed. In this study, both the VADER lexicon (for polarity categories like positive, negative, and neutral) and the NRC lexicon (for specific emotions like joy, anger, and fear) were used. Each lexicon scores the text differently, allowing for a nuanced view of how sentiment is expressed in the data.
Why this is valuable for cultural data:
Our project topic is Literature and Texts – specifically Children's Literature. Sentiment Analysis can reveal how characters and themes are portrayed emotionally across narratives, helping to identify cultural trends and shifts in attitudes within children's literature and film.
Scholars:
Frangidis, P., Georgiou, K., Papadopoulos, S. (2020). Sentiment Analysis on Movie Scripts and Reviews. In: Maglogiannis, I., Iliadis, L., Pimenidis, E. (eds) Artificial Intelligence Applications and Innovations. AIAI 2020. IFIP Advances in Information and Communication Technology, vol 583. Springer, Cham. https://link.springer.com/chapter/10.1007/978-3-030-49161-1_36
Review and Summarize Selected Scholarship:
Bibliographic Information:
Authors: Frangidis, P., Georgiou, K., & Papadopoulos, S.
Title: Sentiment Analysis on Movie Scripts and Reviews: Utilizing Sentiment Scores in Rating Prediction
Publication Venue: AIAI 2020, IFIP AICT
Publication Date: 2020
Computational Method or Cultural Data Type
Method: The study applies Sentiment Analysis by combining lexicons (VADER for sentiment scoring and NRC for emotion
categorization) to movie scripts and reviews, aiming to predict movie ratings based on emotional alignment between scripts and
reviews.
Transformation: Text data is processed to assign sentiment scores, capturing shifts in emotion throughout the movie scripts and
reviews. This transformation allows for the analysis of how emotional tone aligns with audience reception.
Summary of Argument and Use of Computational Method:
Argument: The authors argue that the alignment or divergence between the intended emotional tone of a movie (in the script) and
the audience's emotional response (in reviews) can help predict movie ratings.
Method Application: The study combines sentiment scores from both scripts and reviews to build predictive models. The use of
lexicons helps identify emotional trends and rates emotional resonance, which are then evaluated using machine learning
algorithms to assess predictive accuracy.
Code and Data Availability:
Data: The dataset includes 747 movie scripts and 78,000 reviews, sourced from the IMSDB and Rotten Tomatoes websites.
Code: The study used sentiment analysis libraries such as VADER and NRC but does not provide full code
Interest, Usefulness, and Disciplinary Approach:
Interest: The study is compelling in its approach to comparing intended and perceived sentiment, which has applications in media
studies and consumer behavior.
Disciplinary Approach: This study lies within the field of Computational Social Sciences and Digital Humanities, integrating machine
learning with cultural and emotional analysis.
Potential Use for Group Project:
This study’s approach could help us analyze themes in children’s literature by looking at emotional patterns. Using both VADER and
NRC lexicons, we can track emotions like “trust” or “fear” throughout stories to see if certain themes (like friendship or conflict)
show up with specific emotional tones. This could also reveal shifts in cultural attitudes, such as how bravery or kindness are
portrayed over time. Plus, by mapping emotions in key story moments, we might get insights into the underlying morals and
messages, making sentiment analysis a versatile tool for exploring cultural themes across narratives.
Lucas Washor - Sentiment Analysis and Film Scripts
Define Sentiment Analysis: Sentiment Analysis is a computational method that assesses the emotional tone or polarity (positive, negative, neutral) in text by analyzing words and phrases. It transforms raw text into sentiment scores—typically as numerical values for positive, negative, or neutral tones—which allow us to quantify emotions expressed in a body of work. This scoring helps reveal emotional and cultural perspectives in narratives, as well as patterns in how themes are emotionally framed. In this study, both the VADER lexicon (for polarity categories like positive, negative, and neutral) and the NRC lexicon (for specific emotions like joy, anger, and fear) were used. Each lexicon scores the text differently, allowing for a nuanced view of how sentiment is expressed in the data.
Why this is valuable for cultural data: Our project topic is Literature and Texts – specifically Children's Literature. Sentiment Analysis can reveal how characters and themes are portrayed emotionally across narratives, helping to identify cultural trends and shifts in attitudes within children's literature and film.
Scholars: Frangidis, P., Georgiou, K., Papadopoulos, S. (2020). Sentiment Analysis on Movie Scripts and Reviews. In: Maglogiannis, I., Iliadis, L., Pimenidis, E. (eds) Artificial Intelligence Applications and Innovations. AIAI 2020. IFIP Advances in Information and Communication Technology, vol 583. Springer, Cham. https://link.springer.com/chapter/10.1007/978-3-030-49161-1_36
Review and Summarize Selected Scholarship:
Bibliographic Information: Authors: Frangidis, P., Georgiou, K., & Papadopoulos, S. Title: Sentiment Analysis on Movie Scripts and Reviews: Utilizing Sentiment Scores in Rating Prediction Publication Venue: AIAI 2020, IFIP AICT Publication Date: 2020
Computational Method or Cultural Data Type Method: The study applies Sentiment Analysis by combining lexicons (VADER for sentiment scoring and NRC for emotion categorization) to movie scripts and reviews, aiming to predict movie ratings based on emotional alignment between scripts and reviews. Transformation: Text data is processed to assign sentiment scores, capturing shifts in emotion throughout the movie scripts and reviews. This transformation allows for the analysis of how emotional tone aligns with audience reception.
Summary of Argument and Use of Computational Method: Argument: The authors argue that the alignment or divergence between the intended emotional tone of a movie (in the script) and the audience's emotional response (in reviews) can help predict movie ratings. Method Application: The study combines sentiment scores from both scripts and reviews to build predictive models. The use of lexicons helps identify emotional trends and rates emotional resonance, which are then evaluated using machine learning algorithms to assess predictive accuracy.
Code and Data Availability: Data: The dataset includes 747 movie scripts and 78,000 reviews, sourced from the IMSDB and Rotten Tomatoes websites. Code: The study used sentiment analysis libraries such as VADER and NRC but does not provide full code
Interest, Usefulness, and Disciplinary Approach: Interest: The study is compelling in its approach to comparing intended and perceived sentiment, which has applications in media studies and consumer behavior. Disciplinary Approach: This study lies within the field of Computational Social Sciences and Digital Humanities, integrating machine learning with cultural and emotional analysis.
Potential Use for Group Project: This study’s approach could help us analyze themes in children’s literature by looking at emotional patterns. Using both VADER and NRC lexicons, we can track emotions like “trust” or “fear” throughout stories to see if certain themes (like friendship or conflict) show up with specific emotional tones. This could also reveal shifts in cultural attitudes, such as how bravery or kindness are portrayed over time. Plus, by mapping emotions in key story moments, we might get insights into the underlying morals and messages, making sentiment analysis a versatile tool for exploring cultural themes across narratives.