A machine learning approach for Bengali handwritten vowel character recognition

Recognition of handwritten characters is complex because the different shapes and numbers characters. Many character recognition strategies have been proposed for both English other major dialects. Bengali generally considered fifth most spoken local language in world. It official widely Bangladesh second among 22 posted dialects India. To improve characters, we developed a approach this study using face mapping. quite effective distinguishing The real highlight that results are more efficient than expected with simple machine learning technique. method uses Python library Scikit-Learn, including NumPy, Pandas, Matplotlib, support vector (SVM) classifier. model dataset derived from BanglaLekha isolated training testing part. new shows positive looks promising. showed accuracy up to 94% particular 91% on average all