Story Sentiment Analyzer and Success Predictor

Inspiration

Inspiration came from watching movies and listening to various speakers. I always felt that, the underlying factor on what constitutes a good movie is how well it explores our emotions and how they are balanced out.

What it does

My project provides an analysis on how well a story is balancing the sentiments to captivate their audience. The intent is to provide recommendations to story writers on how to write effectively. It can be used to get recommendations based on type of target audience.

How we built it

GenerativeAI LLM to pass the story text and asked it to generate multiple metrics on emotional sentimental analysis.

Challenges we ran into

Identifying the right balance of emotions and sentiments is hard. Different people have different way of handling the emotions, hence there's no standardized formula for a good story. I had to deep dive on multiple dimensions to understand these aspects.

Accomplishments that we're proud of

The app can recommend if there's an over-indexing on a specific type of emotions and also how it relates to the target audience.

What's next for Story Sentiment Analyzer and Success Predictor

We can scale this idea to any type of story such as College presentation, Business Seminars, Motivational Sessions. Also i want to develop an algorithm which can generate a Success number (out of 100) taking into account the story and typical target audience. Additionally, improving the recommendation engine to provide narrative changes into the story to automatically improve the storytelling experience.

Built With

• amazon-bedrock
• generative-ai