In-depth investigation and use of several machine learning classification techniques for the identification of Gujarati handwritten characters with modifiers are presented in this research article. Gujarati poses specific challenges for precise character identification since it is a complicated script with a wide variety of characters and modifiers. The study's main objective is to assess how well various machine learning algorithms perform when it comes to correctly identifying Gujarati characters when there are modifiers present. The study makes use of a dataset that was gathered from several sources and consists of handwritten Gujarati characters with modifiers. Various machine learning classification techniques, including Naive Bayes, Decision Tree, K-Nearest Neighbours (KNN), Random Forest, Support Vector Machines (SVM), Logistic Regression, Linear Regression, Random Forest, and Random Forest, are optimised for optimal performance.