Portfolio
Hello👋,
Welcome to my portfolio! My name is Saka Wijaya, and I have a strong interest in the field of Data Science and am always eager to learn and develop my skills in this area. Below, you will find an overview of some of the projects I have worked on, showcasing my skills in data analysis, machine learning, and visualization.
Project 1: Analyzing Flight Booking Websites
This project involves analyzing data from flight booking websites to gain insights into user behavior and website performance. The analysis includes data on active users, ticket bookings, customer satisfaction, conversion rate, and error rate. I utilized Python for data processing and cleaning. Additionally, a dashboard has been created to visualize these insights effectively.
Key Metrics Analyzed:
- Active Users: Number of users actively using the flight booking websites.
- Ticket Bookings: Total number of tickets booked through the websites.
- Customer Satisfaction: Rating of user satisfaction based on feedback and reviews.
- Conversion Rate: Percentage of users who completed a booking out of total website visitors.
- Error Rate: Frequency of errors encountered by users during the booking process.
The dashboard provides a comprehensive overview of these metrics and helps in identifying trends and areas for improvement.
Dashboard link here.
Full Project on Analyzing Flight Booking Websites.
Project 2: OWID Covid 19 Analysis
This project involves performing Exploratory Data Analysis (EDA) on the Our World in Data COVID-19 dataset using Python. The analysis includes data exploration, data cleaning, and visualization. Additionally, a dashboard is created using Google Looker Studio to visualize the insights derived from the dataset. The dashboard provides information on which country has the highest total cases, daily cases, the comparison between deaths and total cases, and the comparison between population and vaccination rates in each country.
Dashboard link here.
Full Project on OWID covid 19
Project 3: IMDB Dataset Analysis
- Analyzed IMDB Top 250 movies dataset using Python in Google Colab.
- Cleaned data by converting dates, standardizing currency values, and handling missing values.
- Conducted EDA to visualize rating distributions, release year trends, and budget vs. earnings comparisons.
- Visualized key insights using histograms, line charts, bar charts, and scatter plots.
- Found that high budgets don’t guarantee high earnings, factors like storyline and cast are crucial.
Results
Corelation between world wide gorss and Bugdet films
Full project on IMDB Analysis
Project 4: Heart Disease Classification using Random Forest
- Developed a machine learning model to classify the presence of heart disease using Python.
- Performed data cleaning and preprocessing, including handling missing values and scaling numerical features.
- Using GridSearchCV to find the best hyperparameters for the models.
- Trained and evaluated models, with Random Forest achieving the best performance (Accuracy: 88.3%, Precision: 88.3%, Recall: 88.3%, F1 Score: 88.3%).
- Demonstrated the effectiveness of the Random Forest model in classifying heart disease presence, indicating its suitability for similar tasks.
Results
| Metrics | Score |
|---|---|
| Accuracy | 0.883333 |
| Precision | 0.882961 |
| Recall | 0.883333 |
| F1 Score | 0.882891 |
Full project on Heart Disease Classification using Random Forest
Project 5: Electricity Consumption Forecasting using XGBoostregressor
This project involves forecasting electricity consumption for the next year using machine learning techniques, specifically the XGBoost regressor. The dataset used for this analysis includes various factors such as consumption, production, nuclear energy, wind energy, hydroelectric power, oil and gas consumption, coal consumption, solar energy, and biomass energy.
Results
The XGBoost regressor model used in this analysis achieved a percentage error of 6.1507% in forecasting electricity consumption
Full Project on Electricity Consumption Forecasting
Project 6: Supermarket Sales Analysis
This project involves performing Exploratory Data Analysis (EDA) on the Supermarket Sales dataset using Python. The analysis includes data exploration, data cleaning, and deriving insights from the dataset. With this project, we can determine the number of transactions from January to March, customer payment habits, the number of male and female customers, whether customers are members or not, and identify the target customers for each product line. These insights can be used by the company to optimize inventory management, tailor marketing strategies, enhance customer satisfaction, and ultimately drive sales growth.
Full project on Supermarket Sales