222
The principal goal of this research is to group different kinds of Iris flowers through the use of various machine learning techniques. We aim to develop models which are able to precisely predict species by analyzing the measurement in sepal and petal terms using a well-known dataset from UCI Machine Learning Repository.
Methodology:
- Measurements for three distinct Iris species’ sepal length, sepal width, petal length and width abound here.
- Resolve missing values, feature normalizing, and model training preparation by means of data preprocessing.
- Several classification techniques such as logistic regression, decision tree, k-nearest neighbors, and support vector machines are used.
- Train models on the dataset under hyperparameter adjustments meant to maximize performance.
- Measuring the effectiveness of a model depending on accuracy criteria including accuracy, precision, recall, and F1-score.
Based on the results, the classifier using the Support Vector Machine achieved highest accuracy in identifying different Iris species. This work demonstrates how different machine learning approaches can be useful in solving classification problems as well as clarifies the approach of selecting and evaluating models.
