Workflow

1. Data Collection: I gathered a dataset of movies from IMDb, including metadata such as genre, plot summary, cast, and user ratings.

2. Data Preprocessing: I cleaned the data by handling missing values, normalizing text fields, and encoding categorical variables.

3. Feature Engineering: I created new features based on the existing metadata, such as the number of genres, average user rating, and presence of certain keywords in the plot summary.

4. Model Training: I trained various models using Scikit-learn to predict movie themes based on chosen features.

5. Evaluation: I evaluated the model's performance using metrics such as accuracy, precision, recall, and F1-score.

Workflow

1. Data Collection: We used web scraping to create a dataset of existing TV show scripts.

2. Data Preprocessing: I cleaned the data by by removing punctuation, filtering on what lines are from which persona, and creating a structured format that would be easy to parse.

3. Parsing: I used NLTK to parse the sentences in the scripts, looking at the structure of the sentences and how they were built (classifying word types, and probabilities of word appearances). I then created a model that could generate new sentences based on this structure.

Workflow

1. Data Collection: I gathered several datasets containing information about diabetes patients and their socioeconomic backgrounds.

2. Data Preprocessing: I cleaned the data by handling missing values and normalizing numerical features.

3. Exploratory Data Analysis: I visualized the data to understand its distribution and identify any patterns or outliers.

4. Statistical Testing: I did various statistical tests to determine the significance of the relationships between variables.

Workflow

1. Bot Creation: I created a Discord bot using the Discord.py library and set up the necessary permissions in the discord developer portal.

2. Tag Management: I implemented a bot function for users to set their unique producer tags, which are stored in a json file.

3. Determining Voice Channel: I added functionality to set the voice channel where the bot will play the producer tags.

Workflow

1. Dataset Download: I downloaded and organized the Google Speech Commands v0.02 dataset.

2. Feature Extraction: I made an MFCC extraction pipeline to transform raw audio into compact spectral features with reduced noise suitable for machine learning.

3. Dataset Preparation: I split the data into train, validation, and test splits to feed the model.

4. Model Architecture: I built a Convolutional Neural Network (CNN) specifically designed to classify the MFCC-based feature maps.

6. Training & Evaluation: Finally I trained and evaluated performance using accuracy and a confusion matrix to find the model strengths and weaknesses.

Website Workflow

1. UI/UX Design: I created a modern looking interface with a focus on smooth scrolling and a responsive layout.

2. Frontend Architecture: I built the core structure using HTML.

3. Interactivity & Animations: I used the GSAP library (GreenSock Animation Platform) to handle scroll-triggered animations between sections.

4. Scripts: I wrote JavaScript logic to manage state, scroll calculations, form submissions, and much more.