v1

index.markdown

@@ -1,51 +1,103 @@
-# A Comedy Handbook
+# DATASTORY
 
-Laughter. This great human sentiment that surrounds us from a very early age is a crucial aspect of cinematography and even has its own genre associated – comedies. Do you remember chuckling at the latest *Hangover*? Do you also remember being bored when watching *The Emoji Movie*?
+## A Comedy Handbook
+
+Laughter, the great human sentiment that surrounds us from a very early age, is a crucial aspect of cinematography and even has its own genre: comedies. Do you remember chuckling at *The Hangover*? Do you also remember being bored while watching *The Emoji Movie*?
+
+Both movies are comedies, yet one was much "better" than the other. This project aims to determine what factors make a great comedy. We will explore four main aspects: 
 
-Both movies are comedies, yet one was much “better” than the other. This project aims to determine what factors make for a great comedy. We will explore four main aspects:
 1. What makes a movie popular?
 2. What makes it qualitative?
 3. Are there cultural aspects influencing this?
-4. To what extent are each of these factors unique to our genre of interest?
+4. To what extent are these factors unique to our genre of interest?
 
----
+### Defining Our Objective
 
-## Defining a "Good Comedy"
+**What do we call a good comedy?**
 
-Before we dive into the analysis, we must first define our objective: **What do we call a good comedy?**
+### 1. Sentiment Analysis
 
-### Key Metrics:
-- **Critical Success**: Recognition through awards. *(Responsibility: Arthur)*
-- **Commercial Success**: Measured by box office revenues. *(Responsibility: Arthur)*
-- **Ratings**: Evaluation from experts and non-experts. *(Responsibility: ???)*
-- **Popularity Score**: Quantifying public interest. *(Responsibility: ???)*
-- **Sentiment Score**: Analyzing audience sentiment. *(Responsibility: Theo)*
+#### Keys to Determining Objectively if a Movie is Good
 
----
+Are the ratings of big critic sites accurate? Do they efficiently reflect the impression of a viewer?
 
-## Components of a Good Comedy
+The correlation between sentiment analysis mean scores and user ratings is quite low (<0.4), which prevents drawing conclusions on the accuracy of these factors to validate if a movie is "good." Low correlation and high spread indicate that a rating does not necessarily reflect the sentiment conveyed by the review. 
 
-To identify what makes a comedy successful, we analyze the following components:
+A bias we might expect is more positive sentiment for comedies than for other types of movies. Statistically, while there are significant differences, comedies do not display much higher sentiment scores than other movie types.
 
-1. **Plot and Review Topics**: Understanding recurring themes and topics in reviews. *(Responsibility: Victor)*
-2. **Subgenre**: Categorizing comedies into various subgenres. *(Responsibility: Arthur)*
-3. **Sequels**: Examining the impact of sequels on the success of comedies. *(Responsibility: Caro)*
-4. **Production**: Analyzing the influence of production countries. *(Responsibility: Arthur)*
-5. **Cast**: Investigating whether actors/actresses with extensive comedic portfolios influence movie success. *(Responsibility: Caro)*
-6. **Title Impact**: Exploring the relationship between a movie’s title and its performance. *(Responsibility: Caro)*
+### 2. Critical Success Worldwide (Oscars and Awards)
 
----
+#### Quantifying Critically Acclaimed Success
 
-## Timing Matters: Release Month Impact
+We can quantify critical success for comedies by examining Oscar nominations and wins. Analyzing the percentage of award-winning comedies across countries yields interesting results. For instance:
 
-Finally, we investigate the timing of a movie’s release. Does the release month affect its success? *(Responsibility: Arthur)*
+- Countries like Slovenia, Azerbaijan, Zimbabwe, and Bosnia and Herzegovina show 100% because their Oscar-winning movies were comedies. However, these results stem from a single comedy win.
+- A better contrast comes from analyzing the trade-off between the number of critically acclaimed movies and the percentage that are comedies. Here, Ireland stands out as a strong source for well-received comedies.
 
----
+### 3. Commercial Success Worldwide (Box Office Revenues)
+
+Comedies consistently contribute to box office revenues globally. For example:
+
+- Comedies account for a smaller percentage of the U.S. box office revenues compared to Serbia and Montenegro.
+- However, in absolute terms, the U.S. produces so many movies that it's still more profitable to make a comedy in the U.S. than in smaller markets.
+
+### 4. Ratings: Experts vs. Non-Experts
+
+### 5. Popularity Score
+
+#### Popularity as a Combination of Sales and Reviews
+
+Initially, we hypothesized a direct correlation between audience ratings and box office revenue. The data showed otherwise: there is no consistent relationship. Instead, we found a strong alignment between expert and audience ratings. 
+
+#### Crafting the Popularity Score
+
+A truly successful comedy balances critical acclaim, audience approval, and financial performance. Our "popularity score" combines these elements:
+
+- **Formula:**
+  \[ \text{Popularity Score} = \alpha \cdot \text{Critical Acclaim} + \beta \cdot \text{Audience Ratings} + \gamma \cdot \text{Normalized Box Office Revenue} \]
+
+Where \( \alpha, \beta, \gamma \) are weights derived from regression analysis.
+
+#### Visualizing Success
 
-## The Comedy Recipe
+1. **Expert Ratings vs. Audience Ratings:** Clear positive correlation.
+2. **Audience Ratings vs. Normalized Box Office Revenue:** Certain high-rating films achieve exceptional financial success.
 
-By synthesizing all these factors, we aim to uncover the ultimate recipe for creating a successful comedy. Whether it's the plot, cast, timing, or cultural influences, each element contributes uniquely to the laughter that resonates with audiences worldwide.
+### Components of a Good Comedy
+
+#### 1. Plot and Review Topics
+#### 2. Popular Genres Over Time
+
+We can see the evolution of genres within comedy over decades.
+
+#### 3. Genre Combinations
+#### 4. Cast Impact
+
+#### Analysis: Collaborations Between Actors and Directors
+
+Frequent collaborations, like Adam Sandler with Dennis Dugan or Judd Apatow with Seth Rogen, often yield successful movies.
+
+#### 5. Sequels
+
+Sequels are a staple of the film industry. Do they guarantee success, or merely attempt to cash in on the original's popularity?
+
+#### 6. Production Location
+
+Production location influences comedy success. For instance:
+
+- U.S.: Romantic comedies tend to perform better.
+- France: Drama comedies are more successful.
+
+#### 7. Title Impact
+
+A title serves as a movie's first impression. Great titles like *Superbad* set the tone effectively. Analyzing word frequencies in successful comedy titles reveals common themes.
+
+### Optimal Release Timing
+
+From our analysis, releasing a comedy in June tends to yield the best box office results. Since the 1980s, the proportion of comedy releases has declined compared to their box office revenue share, highlighting their profitability.
+
+---
 
-Let’s find out what makes us laugh!
+### TODO
 
-## References
+- Victor: Finalize with a small prediction model (SLM).