Name: Altay Seipenov
https://drive.google.com/drive/folders/1ikp7TA0ySBkPsiF6As5EGlLrXBMb1ExP?usp=sharing
My ideas for my final project are...
-
Credit Card Approval Prediction A model to predict whether an application gets a green light or a red flag. Classification model development, Handling imbalanced datasets, Model deployment considerations Tools: Python Scikit-learn SVM Data: https://archive.ics.uci.edu/dataset/27/credit+approval https://www.kaggle.com/datasets/rikdifos/credit-card-approval-prediction
-
Uber Data Analysis to gain insights and calculate the price Exploratory Data Analysis (EDA), Data visualization techniques, Advanced data querying and filtering, Feature engineering Tools: Python Pandas Matplotlib/Seaborn Data time Data: https://www.kaggle.com/datasets/fivethirtyeight/uber-pickups-in-new-york-city
I will do Uber Analysis. My execution plan is:
- Week 4: Understanding the Business Model
- I’ll dive deep into understanding Uber’s business model and pinpoint the challenges that can be tackled with data analytics.
- I’ll also familiarize myself with Uber’s Machine Learning Model and Dynamic Pricing Model.
- Week 5: Data Collection and Preprocessing
- I’ll gather all the necessary data for my analysis, which might include details about Uber pickups, ride durations, distances, and pricing.
- I’ll then preprocess this data to make it ready for analysis, which might involve cleaning up the data, handling missing values, and possibly extracting features.
- Week 6: Exploratory Data Analysis
- I’ll conduct an exploratory data analysis on the data I’ve collected. This will help me understand the distribution of data, spot any outliers, and comprehend the relationships between different data points.
- Week 7: Predictive Modelling
- I’ll start constructing a predictive model using the preprocessed data. This will involve selecting a suitable machine learning algorithm, training the model, and fine-tuning it for optimal performance.
- Week 8: Evaluation and Iteration
- I’ll evaluate the predictive model’s performance using various metrics to see how well it’s doing.
- Based on this evaluation, I’ll refine the model to enhance its performance. This might involve adjusting the model parameters, trying out different algorithms, or revisiting the data preprocessing stage to engineer new features.