Linguistic Classification Using Instance-Based Learning

Abstract Traditionally, linguists have organized languages of the world as language families, such Indo-European, Dravidian and Sino-Tibetan. Within Indo-European family, they further into sub-families Germanic, Celtic Indo-Iranian. They do this by looking at similar-sounding words across commonality rules word formation sentence construction. In work, we make use computational approaches that are more scalable. More importantly, contest tree-based structure family models follow, which feel is rather constraining comes in way natural discovery relationships between any two languages. For example, affinity Sanskrit has with Irish, Iranian or English better illustrated using a network model. Similarly, Indian inter-relationships go beyond confines Indo-Aryan divide. To enable languages, paper, instance-based learning techniques to assign labels words. Our approach comprises building corpus then applying clustering construct training set. Following this, vocalized classified making custom linguistic distance metric. We considered seven namely Kannada, Marathi, Punjabi, Hindi, Tamil, Telugu Sanskrit. believe our work potential usher new era linguistics India.KeywordsLinguisticsAryan invasion theoryOut India theorySoundex scoreInstance-based learningKNNDBSCANClusteringClustering coefficient