ADAdasurmonbidet2024 - DISCOVER CHARACTERS' FEATURES BASED ON FIRST NAME

Project Ideas

In film storytelling, a character’s name is rarely just a label; it often serves as a subtle cue to their personality and role in the movie. Our project tackles the intriguing question: "Can a character’s archetype be predicted from their name?" Through this data analysis project, we aim to decode connections between specific name characteristics — such as length, structure, cultural origin, and phonetics — and features including gender, narrative role, movie genre-specific role, and additional properties that we expect to uncover through deeper analysis. A sentiment analysis will also allow us to explore whether these name characteristics and features evoke positive or negative associations, potentially correlating with archetypes such as hero, villain, or more nuanced characters, if time permits.

Research Questions

• What are the key predictors of character traits based on first names?
• Can the number of vowels, consonants, or specific sounds in a name correlate with the character’s role (e.g., hero, villain)?
• How do the length and structure of a character’s first name correlate with their role in the narrative? Do longer or more complex names correspond to more prominent or villainous roles in films?
• Can sentiment analysis of names offer insights into character traits?
• Do films produced in specific regions tend to use names that align with the cultural background of the characters? For example, do American films predominantly feature Anglo-Saxon names for protagonists and villainous roles?

Additional Dataset

We propose integrating a dataset that associates each first name with its ethnicity, which would allow us to make the ethnic origin of the names one of the characteristics of our analysis. This would enable us to, for example, explore the relationship between the origin of a name and the location where the films are produced. The dataset we have identified, "Name Ethnicity," is available on Kaggle.

Project Plans and Methods

Task 0: Data Exploration

We began with data cleaning, removing irrelevant columns and rows, particularly those missing the character name. For preprocessing, we isolated first names by:

• Stripping out prefixes (e.g., “Dr. Alison Parker” becomes “Alison Parker”)
• Eliminate commun names like “taxi” or “waiter” using nltk.corpus.words.
• From the filtered data, restore words identified as valid names using the nltk.corpus.names library, which provides a list of English names.
• Keeping only the first name from full names (e.g., “Alison” from “Alison Parker")

The cleaned data was saved in the cleaned.csv file for streamlined analysis.

Task 1: Present globally the data

Task 2: Feature Extraction for Name Analysis

We aim to analyze gender and ethnicity, as they are the only two fixed character traits that remain independent of contextual factors such as the production country or movie genre. We examined the name characteristics by gender to justify the division of male and female for the rest of the analysis. We analysed several characteristics of the names :

• Structure :
  • length using Python’s len() function
  • first and last letters
  • vowel and consonant count using pandas
  • n-gram (first test, will be done for P3)
• Phonetics (P3, optional) using algorithms like Soundex or Metaphone to explore pronunciation patterns and trends related to names

We proved significant differences in the structure of female/male names.

For P3 : We'll add (using an external dataset: "Name_Ethnicity") an analysis on: Ethnicity. This aims to compare the origins of character names with the country of production of the movie and the roles assigned to these characters, revealing cultural patterns or contrasts in naming and characterization within films. We'll be looking for differences in structure and/or phonetic of name from different ethnicity.

Task 3: Understanding the Data

To better understand the relationships between character names and movies, we started by looking at the data to identify patterns and formulate hypotheses. Our goal was to find trends in character features, such as the distribution of names across different movie genres, countries and ages.

Task 4: Study of Trends Between Name Characteristics and Character Features

We identified statistically significant associations between names and movie genre/country. We plan on identifying trends in the data, especially between name characteristics and character features. We will examine whether certain sounds are tied to specific genres or character roles and explore the influence of filming locations on name origins.

Task 5: Sentiment Analysis

The goal is to analyze the connotations of character names and how they are linked to name characteristics. We will conduct sentiment analysis on sentences featuring these names, identifying structures and adjectives to determine if the names have positive, negative, or neutral connotations. Phonetic analysis of names will further explore whether phonetic traits (like name length or consonant count) correlate with specific roles, such as villains.

Task 6: Predictive Model & Generative List of Heroes/Villains

We aim to develop a “Character ID” model that predicts the most likely characteristics of a movie character based on their first name. The model would generate a character profile, including attributes such as gender, ethnicity, age, movie genre and whether the character is a villain or hero.

Task 7: Create Data Story

Create the website.

Organisation:

• Luca: Task 6
• AInhoa: Task 2 and 4
• Fanny: Task 5 and 7
• Zacharie: Task 4 and 6
• Amaury: Task 2 and 4

Can be rearanged if some Tasks are longer than others.

Proposed Timeline

• 22.11.2023: Task 1 to 3 completely done, work on Task 5
• 29.11.2024: Homework 2
• 06.12.2023: Task 4 done, start Task 6, finalize Task 5
• 13.12.2023: Task 6 and 7
• 20.12.2023: Deadline P3