Project 1

Exploring different types of machine learning algorithms that can be performed on the 1985 automobile dataset, as well as data cleaning and descriptive analytics.

Data Cleaning:

Before we dive into machine learning algorithms, it's essential to clean and preprocess the dataset. Here are some steps we can take:

Handling missing values: Check for missing values in the dataset and decide on a strategy to handle them, such as mean/median imputation or listwise deletion.

Data normalization: Normalize the numerical columns to prevent features with large ranges from dominating the model.

Data transformation: Transform categorical variables into numerical variables using techniques like one-hot encoding or label encoding.

Outlier detection: Identify and handle outliers in the dataset, if any.

Descriptive Analytics algorithms that can be performed on the 1985 automobile dataset:

To perform descriptive analytics, we can calculate various statistics and visualizations to understand the distribution of the data. Here are some examples:

Summary statistics: Calculate mean, median, mode, and standard deviation for numerical columns.

Frequency distributions: Calculate frequency distributions for categorical columns.

Correlation analysis: Calculate correlation coefficients between numerical columns to identify relationships.

Visualizations: Create plots like histograms, box plots, and scatter plots to visualize the distribution of the data.

Machine Learning Algorithms

Here are some machine learning algorithms that can be performed on the 1985 automobile dataset:

Regression algorithms:

Linear Regression: to predict continuous variables like price, horsepower, and fuel efficiency.

Decision Trees: to predict continuous variables and handle non-linear relationships.

Random Forest: to predict continuous variables and handle non-linear relationships with high accuracy.

Classification algorithms:

Logistic Regression: to predict categorical variables like fuel type (diesel/gas) and aspiration (std/turbo).

Decision Trees: to predict categorical variables and handle non-linear relationships.

Random Forest: to predict categorical variables and handle non-linear relationships with high accuracy.

Clustering algorithms:

K-Means: to group similar vehicles based on their characteristics.

Hierarchical Clustering: to group similar vehicles based on their characteristics and visualizes the hierarchy.